@ingglish/ipa 0.1.0

package/dist/index.cjs

@@ -0,0 +1,3905 @@
+"use strict";
+var __defProp = Object.defineProperty;
+var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
+var __getOwnPropNames = Object.getOwnPropertyNames;
+var __hasOwnProp = Object.prototype.hasOwnProperty;
+var __export = (target, all) => {
+  for (var name in all)
+    __defProp(target, name, { get: all[name], enumerable: true });
+};
+var __copyProps = (to, from, except, desc) => {
+  if (from && typeof from === "object" || typeof from === "function") {
+    for (let key of __getOwnPropNames(from))
+      if (!__hasOwnProp.call(to, key) && key !== except)
+        __defProp(to, key, { get: () => from[key], enumerable: !(desc = __getOwnPropDesc(from, key)) || desc.enumerable });
+  }
+  return to;
+};
+var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: true }), mod);
+
+// src/index.ts
+var index_exports = {};
+__export(index_exports, {
+  LANGUAGES: () => LANGUAGES,
+  NOT_FOUND_MARKER: () => NOT_FOUND_MARKER,
+  arpabetPhonemeToIPA: () => arpabetPhonemeToIPA,
+  arpabetToIPARaw: () => arpabetToIPARaw,
+  ipaToArpabetClean: () => ipaToArpabetClean,
+  ipaToIngglish: () => ipaToIngglish,
+  lookupIpa: () => lookupIpa,
+  registerIPA: () => registerIPA,
+  segmentKhmerText: () => segmentKhmerText,
+  translateForeign: () => translateForeign,
+  translateForeignWithMapping: () => translateForeignWithMapping
+});
+module.exports = __toCommonJS(index_exports);
+var import_phonemes4 = require("@ingglish/phonemes");
+
+// src/to-ipa.ts
+var import_phonemes = require("@ingglish/phonemes");
+
+// src/ipa-maps.ts
+var IPA_VOWEL_MAP = {
+  // Monophthongs
+  AA: "\u0251",
+  // father, hot, bother
+  AE: "\xE6",
+  // cat, bat, had
+  AH: "\u028C",
+  // but, cup, son (stressed)
+  AO: "\u0254",
+  // thought, caught, law
+  // Diphthongs
+  AW: "a\u028A",
+  // cow, how, out
+  AY: "a\u026A",
+  // my, eye, time
+  EH: "\u025B",
+  // bed, red, said
+  ER: "\u025D",
+  // bird, her, nurse
+  EY: "e\u026A",
+  // say, day, make
+  IH: "\u026A",
+  // bit, sit, gym
+  IY: "i",
+  // bee, see, machine
+  OW: "o\u028A",
+  // go, show, coat
+  OY: "\u0254\u026A",
+  // boy, toy, coin
+  UH: "\u028A",
+  // book, put, could
+  UW: "u"
+  // too, blue, food
+};
+var IPA_CONSONANT_MAP = {
+  // Stops (plosives)
+  B: "b",
+  // Affricates
+  CH: "t\u0283",
+  // chat, batch
+  D: "d",
+  // Fricatives
+  DH: "\xF0",
+  // the, this (voiced dental)
+  F: "f",
+  G: "\u0261",
+  // Note: IPA uses ɡ (U+0261), not g
+  HH: "h",
+  JH: "d\u0292",
+  // just, edge
+  K: "k",
+  // Liquids
+  L: "l",
+  // Nasals
+  M: "m",
+  N: "n",
+  NG: "\u014B",
+  // sing, thing
+  P: "p",
+  R: "\u0279",
+  // alveolar approximant
+  S: "s",
+  SH: "\u0283",
+  // ship
+  T: "t",
+  TH: "\u03B8",
+  // think (voiceless dental)
+  V: "v",
+  // Glides (semivowels)
+  W: "w",
+  Y: "j",
+  Z: "z",
+  ZH: "\u0292"
+  // measure, beige
+};
+var ARPABET_TO_IPA_MAP = {
+  ...IPA_VOWEL_MAP,
+  ...IPA_CONSONANT_MAP
+};
+var IPA_VARIANT_MAP = {
+  a: "AE",
+  // plain /a/ — maps to "a" (cat) for recognizable foreign word output
+  e: "EH",
+  // plain /e/ — mid front vowel, like "bed"
+  \u0259: "AH0",
+  // schwa (unstressed) — forward map uses ʌ→AH for the stressed variant
+  \u025A: "ER",
+  // r-colored schwa variant — forward map uses ɝ→ER
+  g: "G",
+  // ASCII g — forward map uses ɡ (U+0261)
+  \u026B: "L",
+  // dark l
+  o: "OW",
+  // some IPA uses plain o for goat vowel
+  r: "R"
+  // common variant — forward map uses ɹ (alveolar approximant)
+  // Note: IPA /y/ is the close front rounded vowel (French "tu", German "über").
+  // It's handled in IPA_APPROXIMATION_MAP as y→UW. The consonant /j/ (palatal
+  // approximant) is already mapped via the forward map reversal (j→Y).
+};
+var IPA_APPROXIMATION_MAP = {
+  // --- Open vowels ---
+  \u00E4: "AA",
+  // /ä/ open central — IPA diacritic variant ≈ "father"
+  // --- Open vowel diphthongs (Finnish, etc.) ---
+  // English uses /aɪ/ and /aʊ/ (with plain 'a'), but Finnish/other languages
+  // use /ɑi/ and /ɑu/ (with open back 'ɑ'). Treat as diphthongs, not two vowels.
+  \u00E6i: "AY",
+  // /æi/ — Finnish "päivä" ≈ "my" diphthong (uses near-open front æ)
+  \u0250: "AH",
+  // /ɐ/ near-open central — Portuguese unstressed "a" ≈ "but"
+  \u0251i: "AY",
+  // /ɑi/ — Finnish "taivas" ≈ "my" diphthong
+  \u0251u: "AW",
+  // /ɑu/ — Finnish "sauna" ≈ "cow" diphthong
+  \u0251\u028A: "AW",
+  // /ɑʊ/ — Chinese 好 /xɑʊ/ ≈ "cow" diphthong (uses IPA ʊ not plain u)
+  // Nasal vowels (ɑ̃, ɛ̃, ɔ̃, etc.) are handled in from-ipa.ts by
+  // converting vowel+combining-tilde to vowel+"n" before map lookup.
+  \u0252: "AO",
+  // /ɒ/ open back rounded — British "lot" ≈ "thought"
+  // --- Implosives and other stops ---
+  \u0253: "B",
+  // /ɓ/ voiced bilabial implosive ≈ B
+  \u00E7: "SH",
+  // /ç/ voiceless palatal fricative — "ich" (German) ≈ "sh"
+  // --- Alveolo-palatal (Mandarin, Japanese, Polish) ---
+  \u0255: "SH",
+  // /ɕ/ voiceless alveolo-palatal fricative — Mandarin "xi" ≈ "sh"
+  d\u0291: "JH",
+  // /dʑ/ voiced alveolo-palatal affricate — Japanese "ji" ≈ "j"
+  // --- Retroflex (Hindi, Mandarin) ---
+  \u0256: "D",
+  // /ɖ/ voiced retroflex stop — Hindi ≈ D
+  \u0257: "D",
+  // /ɗ/ voiced alveolar implosive ≈ D
+  // --- Diphthong sequences (non-English vowel pairs) ---
+  // These use standard IPA vowels that don't appear in English diphthongs
+  // (English uses ɪ/ʊ as second element, not i/u/o).
+  ei: "EY",
+  // /ei/ — Finnish "ei", various ≈ "say" diphthong
+  \u0258: "AH0",
+  // /ɘ/ close-mid central ≈ schwa
+  \u025C: "ER",
+  // /ɜ/ open-mid central — non-rhotic "bird" ≈ "er"
+  \u025E: "ER",
+  // /ɞ/ open-mid central rounded ≈ "er"
+  \u0264: "AH",
+  // /ɤ/ close-mid back unrounded — Korean "ㅓ" ≈ "but"
+  \u0262: "G",
+  // /ɢ/ voiced uvular stop ≈ G
+  \u0263: "G",
+  // /ɣ/ voiced velar fricative — Spanish "lago" ≈ G
+  // --- Pharyngeal (Arabic) ---
+  \u0127: "HH",
+  // /ħ/ voiceless pharyngeal fricative — Arabic "ha" ≈ H
+  \u0266: "HH",
+  // /ɦ/ voiced glottal fricative — Korean 합 ≈ H
+  // --- Central/back vowels not in English ---
+  \u0268: "IH",
+  // /ɨ/ close central — Russian "ы" ≈ "bit"
+  \u026C: "L",
+  // /ɬ/ voiceless lateral fricative — Welsh "ll" ≈ L
+  \u026D: "L",
+  // /ɭ/ retroflex lateral ≈ L
+  // --- Laterals ---
+  \u028E: "L Y",
+  // /ʎ/ palatal lateral — Italian "figlio", Spanish "ll" ≈ LY
+  \u0271: "M",
+  // /ɱ/ labiodental nasal ≈ M
+  \u0274: "N",
+  // /ɴ/ uvular nasal — Japanese moraic ん ≈ "n" (not "ng")
+  // --- Nasals ---
+  \u0272: "N Y",
+  // /ɲ/ palatal nasal — Spanish "ñ", Italian "gn" ≈ NY
+  // tɕ and dʑ are handled as two-char sequences below
+  \u0273: "N",
+  // /ɳ/ retroflex nasal ≈ N
+  \u00F8: "UH",
+  // /ø/ close-mid front rounded — "peu" (French), "schön" (German) ≈ "u"
+  \u0153: "AH1",
+  // /œ/ open-mid front rounded — "peur" (French) ≈ "uh"
+  \u0153y: "OY",
+  // /œy/ — Dutch "huis" diphthong ≈ "boy"
+  oi: "OY",
+  // /oi/ — Finnish "koira", Portuguese "coisa" ≈ "boy" diphthong
+  ou: "OW",
+  // /ou/ — Finnish "koulu" ≈ "go" diphthong
+  o\u026F: "OW",
+  // /oɯ/ — Japanese long /oː/ (dict convention: 東京 /toɯkjoɯ/) ≈ "oh"
+  \u0278: "F",
+  // /ɸ/ voiceless bilabial fricative — Japanese "fu" ≈ F
+  q: "K",
+  // /q/ voiceless uvular stop — Arabic "Quran" ≈ K
+  \u0280: "R",
+  // /ʀ/ uvular trill — some German dialects ≈ English R
+  // --- Taps, trills, and approximants (Spanish, Italian, Mandarin, etc.) ---
+  \u027B: "R",
+  // /ɻ/ voiced retroflex approximant — Mandarin "er" (二) ≈ R
+  \u027D: "D",
+  // /ɽ/ retroflex flap — Hindi ≈ D
+  \u027E: "R",
+  // /ɾ/ alveolar tap — Spanish "pero" ≈ R (also like "butter" flap)
+  // --- Uvular consonants (French, German, Arabic) ---
+  \u0281: "R",
+  // /ʁ/ voiced uvular fricative — French/German R ≈ English R
+  \u0282: "SH",
+  // /ʂ/ voiceless retroflex fricative — Mandarin "shi" ≈ "sh"
+  // --- Affricates (two-char sequences matched by the converter) ---
+  t\u0255: "CH",
+  // /tɕ/ voiceless alveolo-palatal affricate — Mandarin "ji", Korean "ㅈ" ≈ "ch"
+  \u0288: "T",
+  // /ʈ/ voiceless retroflex stop — Hindi ≈ T
+  \u0288\u0282: "CH",
+  // /ʈʂ/ voiceless retroflex affricate — Mandarin "zhi" ≈ "ch"
+  // --- Labial-velar ---
+  \u0265: "W",
+  // /ɥ/ labial-palatal approximant — French "lui" ≈ W
+  \u026F: "UH",
+  // /ɯ/ close back unrounded — Japanese "u", Turkish "ı" ≈ "book" (shorter than "oo")
+  \u0270: "W",
+  // /ɰ/ voiced velar approximant — Japanese 川 /kaɰᵝa/ ≈ W
+  \u028B: "V",
+  // /ʋ/ labiodental approximant — Hindi, Dutch ≈ V
+  // --- Velar/palatal fricatives (German, Mandarin, etc.) ---
+  x: "HH",
+  // /x/ voiceless velar fricative — default H (override to K for German)
+  // --- Front rounded vowels (French, German, Turkish, etc.) ---
+  y: "UW",
+  // /y/ close front rounded — "tu" (French) ≈ "too"
+  \u028F: "UH",
+  // /ʏ/ near-close front rounded — "Glück" (German) ≈ "book"
+  \u0290: "ZH",
+  // /ʐ/ voiced retroflex fricative — Mandarin "ri" ≈ "zh"
+  \u0291: "ZH",
+  // /ʑ/ voiced alveolo-palatal fricative ≈ "zh"
+  // --- Glottal ---
+  \u0294: "",
+  // /ʔ/ glottal stop — often silent in approximation
+  // Note: length mark ː and combining diacritics are stripped in from-ipa.ts
+  // before lookup. Nasal vowels (◌̃) are converted to vowel+n there.
+  // --- Dental fricatives already in English ---
+  // θ and ð are in the main map
+  \u0295: "AH",
+  // /ʕ/ voiced pharyngeal fricative — Arabic "ain" ≈ "uh" (rough)
+  // --- Bilabial fricatives (Spanish) ---
+  \u03B2: "V",
+  // /β/ voiced bilabial fricative — Spanish "b" between vowels ≈ V
+  \u03C7: "HH"
+  // /χ/ voiceless uvular fricative — default H (override to K for German)
+};
+var IPA_TO_ARPABET_MAP = {
+  ...IPA_APPROXIMATION_MAP,
+  ...Object.fromEntries(Object.entries(ARPABET_TO_IPA_MAP).map(([arpabet, ipa]) => [ipa, arpabet])),
+  ...IPA_VARIANT_MAP
+};
+var IPA_LANGUAGE_OVERRIDES = {
+  // German: /x/ and /χ/ are the "ach-Laut" — English speakers say "bahk" for Bach
+  de: { x: "K", \u03C7: "K" },
+  // Dutch: /ɣ/ is a breathy G (closer to H than hard G) — "goed" ≈ "hood" not "good"
+  nl: { \u0263: "HH" }
+};
+
+// src/to-ipa.ts
+var WORD_JOINER = "\u2060";
+var STRESS_MARKERS = {
+  0: "",
+  1: WORD_JOINER + "\u02C8" + WORD_JOINER,
+  2: WORD_JOINER + "\u02CC" + WORD_JOINER
+};
+function arpabetPhonemeToIPA(phoneme) {
+  const base = (0, import_phonemes.stripStress)(phoneme);
+  const stress = (0, import_phonemes.getStress)(phoneme);
+  const ipa = ARPABET_TO_IPA_MAP[base];
+  if (ipa === void 0) {
+    return phoneme.toLowerCase();
+  }
+  if (base === "AH" && stress === 0) {
+    return "\u0259";
+  }
+  if (stress === 1 || stress === 2) {
+    return STRESS_MARKERS[stress] + ipa;
+  }
+  return ipa;
+}
+function arpabetToIPA(arpabet) {
+  const ipaSegments = [];
+  const stressPositions = [];
+  for (let i = 0; i < arpabet.length; i++) {
+    const symbol = arpabet[i];
+    const base = (0, import_phonemes.stripStress)(symbol);
+    const stress = (0, import_phonemes.getStress)(symbol);
+    const ipa = ARPABET_TO_IPA_MAP[base];
+    if (ipa === void 0) {
+      ipaSegments.push(symbol.toLowerCase());
+      continue;
+    }
+    if (base === "AH" && stress === 0) {
+      ipaSegments.push("\u0259");
+      continue;
+    }
+    if (stress === 1 || stress === 2) {
+      const marker = stress === 1 ? WORD_JOINER + "\u02C8" + WORD_JOINER : WORD_JOINER + "\u02CC" + WORD_JOINER;
+      let onsetIndex = ipaSegments.length;
+      if (i > 0) {
+        let j = i - 1;
+        const consonants = [];
+        while (j >= 0 && !(0, import_phonemes.isVowel)(arpabet[j])) {
+          consonants.push((0, import_phonemes.stripStress)(arpabet[j]));
+          j--;
+        }
+        consonants.reverse();
+        if (consonants.length > 0) {
+          const onsetStartInCluster = (0, import_phonemes.findOnsetStart)(consonants);
+          onsetIndex = j + 1 + onsetStartInCluster;
+        }
+      }
+      stressPositions.push({ index: onsetIndex, marker });
+    }
+    ipaSegments.push(ipa);
+  }
+  const sortedStress = stressPositions.toSorted((a, b) => a.index - b.index);
+  const result = [];
+  let stressIdx = 0;
+  for (let i = 0; i <= ipaSegments.length; i++) {
+    while (stressIdx < sortedStress.length && sortedStress[stressIdx].index === i) {
+      result.push(sortedStress[stressIdx].marker);
+      stressIdx++;
+    }
+    if (i < ipaSegments.length) {
+      result.push(ipaSegments[i]);
+    }
+  }
+  return `/${result.join("")}/`;
+}
+function arpabetToIPARaw(arpabet) {
+  const full = arpabetToIPA(arpabet);
+  return full.slice(1, -1);
+}
+
+// src/foreign.ts
+var import_normalize = require("@ingglish/normalize");
+var import_phonemes3 = require("@ingglish/phonemes");
+
+// src/from-ipa.ts
+var import_phonemes2 = require("@ingglish/phonemes");
+var ARPABET_VOWELS = new Set(Object.keys(IPA_VOWEL_MAP));
+var NASAL_VOWEL_RE = /([aeiouɑɛɔəɐɒæøœʌɝɚɘɜɞɤʏʊɪɨɯy])\u0303/g;
+var MODIFIER_RE = /[\u02B0\u02D0\u02D1\u02E5-\u02E9\u0303\u1D5D]/g;
+var COMBINING_NON_CEDILLA_RE = /(?!\u0327)\p{Mn}/gu;
+var mergedMapCache = /* @__PURE__ */ new Map();
+function ipaToArpabet(ipa, overrides) {
+  const normalized = ipa.normalize("NFD");
+  const denasalized = normalized.replaceAll(NASAL_VOWEL_RE, "$1n");
+  const stripped = denasalized.replaceAll(MODIFIER_RE, "");
+  const stripped2 = stripped.replaceAll(COMBINING_NON_CEDILLA_RE, "");
+  const clean = stripped2.normalize("NFC");
+  let map = IPA_TO_ARPABET_MAP;
+  if (overrides) {
+    let cached = mergedMapCache.get(overrides);
+    if (!cached) {
+      cached = { ...IPA_TO_ARPABET_MAP, ...overrides };
+      mergedMapCache.set(overrides, cached);
+    }
+    map = cached;
+  }
+  const result = [];
+  let pendingStress = null;
+  let i = 0;
+  const push = (phoneme) => {
+    const base = (0, import_phonemes2.stripStress)(phoneme);
+    if (ARPABET_VOWELS.has(base) && pendingStress !== null) {
+      result.push(base + String(pendingStress));
+      pendingStress = null;
+    } else {
+      result.push(phoneme);
+    }
+  };
+  while (i < clean.length) {
+    const ch = clean[i];
+    if (ch === "\u02C8") {
+      pendingStress = 1;
+      i++;
+      continue;
+    }
+    if (ch === "\u02CC") {
+      pendingStress = 2;
+      i++;
+      continue;
+    }
+    if (i + 1 < clean.length) {
+      const twoChar = clean.slice(i, i + 2);
+      const twoCharArpabet = map[twoChar];
+      if (twoCharArpabet !== void 0) {
+        if (twoCharArpabet.includes(" ")) {
+          for (const p of twoCharArpabet.split(" ")) {
+            push(p);
+          }
+        } else {
+          push(twoCharArpabet);
+        }
+        i += 2;
+        continue;
+      }
+    }
+    const oneCharArpabet = map[ch];
+    if (oneCharArpabet !== void 0) {
+      if (oneCharArpabet.includes(" ")) {
+        for (const p of oneCharArpabet.split(" ")) {
+          push(p);
+        }
+      } else {
+        push(oneCharArpabet);
+      }
+    }
+    i++;
+  }
+  const deduped = [];
+  for (const phoneme of result) {
+    if (phoneme !== deduped.at(-1)) {
+      deduped.push(phoneme);
+    }
+  }
+  return deduped;
+}
+function ipaToArpabetClean(ipa) {
+  const arpabet = ipaToArpabet(ipa).map((p) => (0, import_phonemes2.stripStress)(p));
+  return arpabet.length > 0 ? arpabet : null;
+}
+
+// src/g2p.ts
+var G2P_CONVERTERS = {
+  eo: esperantoG2P,
+  fi: finnishG2P,
+  ma: malayG2P,
+  sw: swahiliG2P
+};
+var IPA_VOWELS = new Set("aeiou\u0251\xE6\xF8y\u025B\u0254\u0259");
+function addFirstSyllableStress(ipa) {
+  return ipa ? "\u02C8" + ipa : ipa;
+}
+function addPenultimateStress(ipa) {
+  const vowelPositions = [];
+  for (const [i, ch] of Array.from(ipa).entries()) {
+    if (IPA_VOWELS.has(ch)) {
+      vowelPositions.push(i);
+    }
+  }
+  if (vowelPositions.length <= 1) {
+    return ipa ? "\u02C8" + ipa : ipa;
+  }
+  const stressPos = vowelPositions.at(-2);
+  let onset = stressPos;
+  while (onset > 0 && !IPA_VOWELS.has(ipa[onset - 1]) && ipa[onset - 1] !== "\u02D0") {
+    onset--;
+  }
+  return ipa.slice(0, onset) + "\u02C8" + ipa.slice(onset);
+}
+function applyRules(word, rules) {
+  const normalized = word.normalize("NFC");
+  let result = "";
+  let i = 0;
+  while (i < normalized.length) {
+    let matched = false;
+    for (const [grapheme, phoneme] of rules) {
+      if (normalized.startsWith(grapheme, i)) {
+        result += phoneme;
+        i += grapheme.length;
+        matched = true;
+        break;
+      }
+    }
+    if (!matched) {
+      i++;
+    }
+  }
+  return result;
+}
+var FINNISH_RULES = [
+  // Digraphs (must precede single-char rules)
+  ["nk", "\u014Bk"],
+  ["ng", "\u014B\u02D0"],
+  // Long vowels (double letters → vowel + length mark)
+  ["aa", "\u0251\u02D0"],
+  ["ee", "e\u02D0"],
+  ["ii", "i\u02D0"],
+  ["oo", "o\u02D0"],
+  ["uu", "u\u02D0"],
+  ["yy", "y\u02D0"],
+  ["\xE4\xE4", "\xE6\u02D0"],
+  ["\xF6\xF6", "\xF8\u02D0"],
+  // Geminate consonants
+  ["pp", "p\u02D0"],
+  ["tt", "t\u02D0"],
+  ["kk", "k\u02D0"],
+  ["mm", "m\u02D0"],
+  ["nn", "n\u02D0"],
+  ["ll", "l\u02D0"],
+  ["rr", "r\u02D0"],
+  ["ss", "s\u02D0"],
+  // Single vowels
+  ["a", "\u0251"],
+  ["e", "e"],
+  ["i", "i"],
+  ["o", "o"],
+  ["u", "u"],
+  ["y", "y"],
+  ["\xE4", "\xE6"],
+  ["\xF6", "\xF8"],
+  // Single consonants
+  ["b", "b"],
+  ["d", "d"],
+  ["f", "f"],
+  ["g", "\u0261"],
+  ["h", "h"],
+  ["j", "j"],
+  ["k", "k"],
+  ["l", "l"],
+  ["m", "m"],
+  ["n", "n"],
+  ["p", "p"],
+  ["r", "r"],
+  ["s", "s"],
+  ["t", "t"],
+  ["v", "\u028B"],
+  ["w", "\u028B"],
+  ["z", "ts"]
+];
+function finnishG2P(word) {
+  return addFirstSyllableStress(applyRules(word, FINNISH_RULES));
+}
+var ESPERANTO_RULES = [
+  // Special Esperanto characters (must precede base letters)
+  ["\u0109", "t\u0283"],
+  ["\u011D", "d\u0292"],
+  ["\u0125", "x"],
+  ["\u0135", "\u0292"],
+  ["\u015D", "\u0283"],
+  ["\u016D", "w"],
+  ["c", "ts"],
+  // Vowels
+  ["a", "a"],
+  ["e", "e"],
+  ["i", "i"],
+  ["o", "o"],
+  ["u", "u"],
+  // Consonants
+  ["b", "b"],
+  ["d", "d"],
+  ["f", "f"],
+  ["g", "\u0261"],
+  ["h", "h"],
+  ["j", "j"],
+  ["k", "k"],
+  ["l", "l"],
+  ["m", "m"],
+  ["n", "n"],
+  ["p", "p"],
+  ["r", "r"],
+  ["s", "s"],
+  ["t", "t"],
+  ["v", "v"],
+  ["z", "z"]
+];
+function esperantoG2P(word) {
+  return addPenultimateStress(applyRules(word, ESPERANTO_RULES));
+}
+var SWAHILI_RULES = [
+  // Trigraph (must precede digraph 'ng')
+  ["ng'", "\u014B"],
+  // Digraphs
+  ["ch", "t\u0283"],
+  ["dh", "\xF0"],
+  ["gh", "\u0263"],
+  ["ng", "\u014B\u0261"],
+  ["nj", "nd\u0292"],
+  ["ny", "\u0272"],
+  ["sh", "\u0283"],
+  ["th", "\u03B8"],
+  // Vowels
+  ["a", "a"],
+  ["e", "\u025B"],
+  ["i", "i"],
+  ["o", "\u0254"],
+  ["u", "u"],
+  // Consonants
+  ["b", "b"],
+  ["d", "d"],
+  ["f", "f"],
+  ["g", "\u0261"],
+  ["h", "h"],
+  ["j", "d\u0292"],
+  ["k", "k"],
+  ["l", "l"],
+  ["m", "m"],
+  ["n", "n"],
+  ["p", "p"],
+  ["r", "\u027E"],
+  ["s", "s"],
+  ["t", "t"],
+  ["v", "v"],
+  ["w", "w"],
+  ["y", "j"],
+  ["z", "z"]
+];
+function swahiliG2P(word) {
+  return addPenultimateStress(applyRules(word, SWAHILI_RULES));
+}
+var MALAY_RULES = [
+  // Digraphs
+  ["gh", "\u0263"],
+  ["kh", "x"],
+  ["ng", "\u014B"],
+  ["ny", "\u0272"],
+  ["sy", "\u0283"],
+  // Vowels
+  ["a", "a"],
+  ["e", "\u0259"],
+  ["i", "i"],
+  ["o", "o"],
+  ["u", "u"],
+  // Consonants
+  ["b", "b"],
+  ["c", "t\u0283"],
+  ["d", "d"],
+  ["f", "f"],
+  ["g", "\u0261"],
+  ["h", "h"],
+  ["j", "d\u0292"],
+  ["k", "k"],
+  ["l", "l"],
+  ["m", "m"],
+  ["n", "n"],
+  ["p", "p"],
+  ["r", "\u027E"],
+  ["s", "s"],
+  ["t", "t"],
+  ["v", "v"],
+  ["w", "w"],
+  ["y", "j"],
+  ["z", "z"]
+];
+function addMalayStress(ipa) {
+  const vowelPositions = [];
+  for (const [i, ch] of Array.from(ipa).entries()) {
+    if (IPA_VOWELS.has(ch)) {
+      vowelPositions.push(i);
+    }
+  }
+  if (vowelPositions.length <= 1) {
+    return ipa ? "\u02C8" + ipa : ipa;
+  }
+  const penultPos = vowelPositions.at(-2);
+  const stressPos = ipa[penultPos] === "\u0259" ? vowelPositions.at(-1) : penultPos;
+  let onset = stressPos;
+  while (onset > 0 && !IPA_VOWELS.has(ipa[onset - 1]) && ipa[onset - 1] !== "\u02D0") {
+    onset--;
+  }
+  return ipa.slice(0, onset) + "\u02C8" + ipa.slice(onset);
+}
+function malayG2P(word) {
+  return addMalayStress(applyRules(word, MALAY_RULES));
+}
+
+// src/lemmatizers.ts
+var LEMMATIZERS = {
+  eo: lemmatizeEo,
+  fa: lemmatizeFa,
+  fi: lemmatizeFi,
+  ma: lemmatizeMa,
+  nb: lemmatizeNb,
+  ro: lemmatizeRo,
+  sv: lemmatizeSv,
+  sw: lemmatizeSw
+};
+function tryLookup(dict, ...candidates) {
+  for (const c of candidates) {
+    if (c && dict[c]) {
+      return dict[c];
+    }
+  }
+  return void 0;
+}
+var SV_SUFFIXES = [
+  // 4+ char suffixes
+  ["erna", ["", "e"]],
+  ["orna", ["", "a"]],
+  ["ande", ["", "a"]],
+  ["ende", ["", "a"]],
+  ["aste", [""]],
+  // 3 char
+  ["ade", ["", "a"]],
+  ["igt", ["ig"]],
+  // 2 char
+  ["en", [""]],
+  ["et", ["", "e"]],
+  ["an", ["", "a"]],
+  ["ar", [""]],
+  ["er", ["", "a"]],
+  ["de", ["", "a"]],
+  ["te", ["", "a"]],
+  // 1 char
+  ["a", [""]],
+  ["t", ["", "a"]],
+  ["s", [""]],
+  ["r", ["", "a"]],
+  ["n", [""]]
+];
+function lemmatizeSv(dict, word) {
+  for (const [suffix, replacements] of SV_SUFFIXES) {
+    if (word.length > suffix.length && word.endsWith(suffix)) {
+      const stem = word.slice(0, -suffix.length);
+      const candidates = replacements.map((r) => stem + r);
+      const ipa = tryLookup(dict, ...candidates);
+      if (ipa) {
+        return ipa;
+      }
+    }
+  }
+  if (word.endsWith("s") && word.length > 2) {
+    const inner = word.slice(0, -1);
+    if (dict[inner]) {
+      return dict[inner];
+    }
+    return lemmatizeSv(dict, inner);
+  }
+  return void 0;
+}
+var RO_SUFFIXES = [
+  // 4+ char
+  ["ului", [""]],
+  ["ilor", [""]],
+  ["e\u0219te", ["", "i"]],
+  // 3 char
+  ["ele", ["", "\u0103"]],
+  ["uri", [""]],
+  ["eau", ["", "i", "ea"]],
+  // 2 char
+  ["ul", [""]],
+  ["ii", ["", "ie", "iu"]],
+  ["ea", ["", "e"]],
+  ["\u021Bi", ["t"]],
+  ["c\u0103", ["c", "ca"]],
+  // 1 char
+  ["a", ["", "\u0103"]],
+  ["e", ["", "\u0103"]],
+  ["i", ["", "e"]]
+];
+var RO_PREFIX_RESTORE = [
+  ["n", "\xEEn"],
+  ["l", "\xEEl"],
+  ["m", "m\u0103"]
+];
+function lemmatizeRo(dict, word) {
+  for (const [suffix, replacements] of RO_SUFFIXES) {
+    if (word.length > suffix.length && word.endsWith(suffix)) {
+      const stem = word.slice(0, -suffix.length);
+      const candidates = replacements.map((r) => stem + r);
+      const ipa = tryLookup(dict, ...candidates);
+      if (ipa) {
+        return ipa;
+      }
+    }
+  }
+  for (const [prefix, restored] of RO_PREFIX_RESTORE) {
+    if (word.startsWith(prefix)) {
+      const remainder = restored + word.slice(prefix.length);
+      if (dict[remainder]) {
+        return dict[remainder];
+      }
+    }
+  }
+  const withI = "\xEE" + word;
+  if (dict[withI]) {
+    return dict[withI];
+  }
+  return void 0;
+}
+var SW_VERB_PREFIXES = [
+  // 5+ char
+  "hatuku",
+  "hawaku",
+  "haiku",
+  "hatua",
+  "hatui",
+  // 4 char
+  "wali",
+  "tuli",
+  "nili",
+  "aali",
+  "wame",
+  "tume",
+  "nime",
+  "ame",
+  "wana",
+  "tuna",
+  "nina",
+  "ana",
+  "wata",
+  "tuta",
+  "nita",
+  "ata",
+  "yame",
+  "yata",
+  "yana",
+  "yali",
+  "kime",
+  "kita",
+  "kina",
+  "kili",
+  "lime",
+  "lita",
+  "lina",
+  "lili",
+  "vime",
+  "vita",
+  "vina",
+  "vili",
+  "zime",
+  "zita",
+  "zina",
+  "zili",
+  "haku",
+  "hatu",
+  "hani",
+  "hawa",
+  // 3 char
+  "ali",
+  "uli",
+  "ume",
+  "una",
+  "uta",
+  "tua",
+  "tui",
+  "wai",
+  "wal",
+  "iku",
+  "ina",
+  "hue",
+  "huj",
+  "hui",
+  "yat",
+  "yam",
+  "yan",
+  "kum",
+  "kui",
+  "kua",
+  // 2 char
+  "wa",
+  "tu",
+  "ni",
+  "li",
+  "ki",
+  "vi",
+  "zi",
+  "ya",
+  "ku",
+  "hu"
+];
+var SW_DERIV_SUFFIXES = [
+  ["ika", ["a"]],
+  ["isha", ["a"]],
+  ["ana", ["a"]],
+  ["wa", ["a"]],
+  ["ia", ["a"]],
+  ["ika", ["a", "ea"]]
+];
+function lemmatizeEo(dict, word) {
+  let w = word;
+  if (w.endsWith("n") && w.length > 2) {
+    const stripped = w.slice(0, -1);
+    if (dict[stripped]) {
+      return dict[stripped];
+    }
+    w = stripped;
+  }
+  if (w.endsWith("j") && w.length > 2) {
+    const stripped = w.slice(0, -1);
+    if (dict[stripped]) {
+      return dict[stripped];
+    }
+    w = stripped;
+  }
+  if (dict[w]) {
+    return dict[w];
+  }
+  for (const ending of ["as", "is", "os", "us"]) {
+    if (w.endsWith(ending) && w.length > ending.length + 1) {
+      const stem = w.slice(0, -ending.length);
+      const ipa = tryLookup(dict, stem + "i", stem + "o", stem);
+      if (ipa) {
+        return ipa;
+      }
+    }
+  }
+  if (w.endsWith("u") && w.length > 2) {
+    const stem = w.slice(0, -1);
+    const ipa = tryLookup(dict, stem + "i", stem + "o", stem);
+    if (ipa) {
+      return ipa;
+    }
+  }
+  for (const suffix of [
+    "anta",
+    "inta",
+    "onta",
+    "ata",
+    "ita",
+    "ota",
+    "ante",
+    "inte",
+    "onte"
+  ]) {
+    if (w.endsWith(suffix) && w.length > suffix.length + 1) {
+      const stem = w.slice(0, -suffix.length);
+      const ipa = tryLookup(dict, stem + "i", stem + "o", stem);
+      if (ipa) {
+        return ipa;
+      }
+    }
+  }
+  if (w.endsWith("e") && w.length > 2) {
+    const stem = w.slice(0, -1);
+    const ipa = tryLookup(dict, stem + "o", stem + "a", stem + "i", stem);
+    if (ipa) {
+      return ipa;
+    }
+  }
+  for (const [suffix, replacements] of [
+    ["isto", ["o", "i", ""]],
+    ["ejo", ["o", "i", ""]],
+    ["ilo", ["o", "i", ""]],
+    ["eco", ["o", "a", ""]],
+    ["ado", ["o", "i", ""]],
+    ["igo", ["o", "a", ""]],
+    ["i\u011Do", ["o", "a", ""]]
+  ]) {
+    if (w.endsWith(suffix) && w.length > suffix.length + 1) {
+      const stem = w.slice(0, -suffix.length);
+      const ipa = tryLookup(dict, ...replacements.map((r) => stem + r));
+      if (ipa) {
+        return ipa;
+      }
+    }
+  }
+  for (const prefix of ["mal", "ek", "re", "ne", "sen"]) {
+    if (w.startsWith(prefix) && w.length > prefix.length + 1) {
+      const remainder = w.slice(prefix.length);
+      if (dict[remainder]) {
+        return dict[remainder];
+      }
+      const inner = lemmatizeEo(dict, remainder);
+      if (inner) {
+        return inner;
+      }
+    }
+  }
+  return void 0;
+}
+function lemmatizeSw(dict, word) {
+  for (const prefix of SW_VERB_PREFIXES) {
+    if (word.length > prefix.length + 1 && word.startsWith(prefix)) {
+      const remainder = word.slice(prefix.length);
+      if (dict[remainder]) {
+        return dict[remainder];
+      }
+      const kuForm = "ku" + remainder;
+      if (dict[kuForm]) {
+        return dict[kuForm];
+      }
+      for (const [suffix, replacements] of SW_DERIV_SUFFIXES) {
+        if (remainder.length > suffix.length && remainder.endsWith(suffix)) {
+          const stem = remainder.slice(0, -suffix.length);
+          for (const r of replacements) {
+            const candidate = stem + r;
+            if (dict[candidate]) {
+              return dict[candidate];
+            }
+            if (dict["ku" + candidate]) {
+              return dict["ku" + candidate];
+            }
+          }
+        }
+      }
+    }
+  }
+  for (const [suffix, replacements] of SW_DERIV_SUFFIXES) {
+    if (word.length > suffix.length && word.endsWith(suffix)) {
+      const stem = word.slice(0, -suffix.length);
+      for (const r of replacements) {
+        const candidate = stem + r;
+        if (dict[candidate]) {
+          return dict[candidate];
+        }
+      }
+    }
+  }
+  return void 0;
+}
+var FI_SUFFIXES = [
+  // Possessive + case combinations
+  ["ssani", ["", "nen"]],
+  ["ss\xE4ni", ["", "nen"]],
+  ["llani", ["", "nen"]],
+  ["ll\xE4ni", ["", "nen"]],
+  ["stani", ["", "nen"]],
+  ["st\xE4ni", ["", "nen"]],
+  ["ssaan", ["", "nen"]],
+  ["ss\xE4\xE4n", ["", "nen"]],
+  // Plural case endings (4+ chars)
+  ["issa", ["", "a"]],
+  ["iss\xE4", ["", "\xE4"]],
+  ["illa", ["", "a"]],
+  ["ill\xE4", ["", "\xE4"]],
+  ["ista", ["", "a"]],
+  ["ist\xE4", ["", "\xE4"]],
+  ["ihin", ["", "i"]],
+  ["ojen", ["o"]],
+  ["ujen", ["u"]],
+  ["yjen", ["y"]],
+  ["iden", ["i"]],
+  ["jen", [""]],
+  // Inessive -ssa/-ssä
+  ["ssa", ["", "s"]],
+  ["ss\xE4", ["", "s"]],
+  // Elative -sta/-stä
+  ["sta", ["", "s"]],
+  ["st\xE4", ["", "s"]],
+  // Adessive -lla/-llä
+  ["lla", ["", "a"]],
+  ["ll\xE4", ["", "\xE4"]],
+  // Ablative -lta/-ltä
+  ["lta", ["", "a"]],
+  ["lt\xE4", ["", "\xE4"]],
+  // Allative -lle
+  ["lle", ["", "i"]],
+  // Essive -na/-nä
+  ["na", ["", "nen"]],
+  ["n\xE4", ["", "nen"]],
+  // Translative -ksi
+  ["ksi", ["", "si"]],
+  // Possessive -ni, -si, -nsa/-nsä, -mme, -nne
+  ["nsa", [""]],
+  ["ns\xE4", [""]],
+  ["mme", [""]],
+  ["nne", [""]],
+  ["ni", ["", "n"]],
+  ["si", ["", "s"]],
+  // Partitive -a/-ä, -ta/-tä, -tta/-ttä
+  ["tta", [""]],
+  ["tt\xE4", [""]],
+  ["ta", ["", "nen"]],
+  ["t\xE4", ["", "nen"]],
+  // Genitive -n, plural -t
+  ["en", ["", "i"]],
+  ["ot", ["o"]],
+  ["ut", ["u"]],
+  ["yt", ["y"]],
+  ["\xE4t", ["\xE4"]],
+  ["at", ["a"]],
+  ["et", ["e", "i"]],
+  // Verb past -i
+  ["oi", ["o", "oa"]],
+  ["ui", ["u", "ua"]],
+  // General fallbacks
+  ["a", [""]],
+  ["\xE4", [""]],
+  ["n", [""]],
+  ["t", [""]]
+];
+var FI_VERB_SUFFIXES = [
+  // Archaic Kalevala -(tt)elevi/-(tt)avi patterns
+  ["ttelevi", ["tella", "della"]],
+  ["televi", ["tella", "della"]],
+  ["ttavi", ["ttaa", "t\xE4\xE4"]],
+  ["ttevi", ["tt\xE4\xE4", "ttaa"]],
+  ["elevi", ["ella", "ell\xE4"]],
+  ["alevi", ["alla", "all\xE4"]],
+  ["evi", ["", "a", "\xE4"]],
+  ["avi", ["", "a", "aa"]],
+  ["ovi", ["", "o", "oa"]],
+  ["uvi", ["", "u", "ua"]],
+  // Past participle -nut/-nyt, -neet
+  ["neet", ["", "a", "\xE4"]],
+  ["nut", ["", "a", "da"]],
+  ["nyt", ["", "\xE4", "d\xE4"]],
+  // Present participle -va/-vä
+  ["va", ["", "a"]],
+  ["v\xE4", ["", "\xE4"]],
+  // Past tense 3rd person
+  ["tui", ["tua", "ty\xE4"]],
+  ["lui", ["la", "l\xE4"]],
+  // Conditional
+  ["isi", ["", "a", "\xE4"]],
+  // Agent noun -ja/-jä
+  ["ja", ["", "a"]],
+  ["j\xE4", ["", "\xE4"]]
+];
+function applyFiGradation(stem) {
+  if (stem.endsWith("nt")) {
+    return stem.slice(0, -2) + "nn";
+  }
+  if (stem.endsWith("lt")) {
+    return stem.slice(0, -2) + "ll";
+  }
+  if (stem.endsWith("rt")) {
+    return stem.slice(0, -2) + "rr";
+  }
+  if (stem.endsWith("nk")) {
+    return stem.slice(0, -2) + "ng";
+  }
+  if (stem.endsWith("mp")) {
+    return stem.slice(0, -2) + "mm";
+  }
+  if (stem.endsWith("lk")) {
+    return stem.slice(0, -2) + "l";
+  }
+  if (stem.endsWith("rk")) {
+    return stem.slice(0, -2) + "r";
+  }
+  if (stem.endsWith("hk")) {
+    return stem.slice(0, -2) + "h";
+  }
+  return stem;
+}
+function applyFiStrengthening(stem) {
+  if (stem.endsWith("nn")) {
+    return stem.slice(0, -2) + "nt";
+  }
+  if (stem.endsWith("ll")) {
+    return stem.slice(0, -2) + "lt";
+  }
+  if (stem.endsWith("rr")) {
+    return stem.slice(0, -2) + "rt";
+  }
+  if (stem.endsWith("ng")) {
+    return stem.slice(0, -2) + "nk";
+  }
+  if (stem.endsWith("mm")) {
+    return stem.slice(0, -2) + "mp";
+  }
+  return stem;
+}
+function lemmatizeFi(dict, word) {
+  for (const [suffix, replacements] of FI_VERB_SUFFIXES) {
+    if (word.length > suffix.length + 1 && word.endsWith(suffix)) {
+      const stem = word.slice(0, -suffix.length);
+      const candidates = replacements.map((r) => stem + r);
+      const ipa = tryLookup(dict, ...candidates);
+      if (ipa) {
+        return ipa;
+      }
+    }
+  }
+  for (const [suffix, replacements] of FI_SUFFIXES) {
+    if (word.length > suffix.length + 1 && word.endsWith(suffix)) {
+      const stem = word.slice(0, -suffix.length);
+      const candidates = replacements.map((r) => stem + r);
+      const ipa = tryLookup(dict, ...candidates);
+      if (ipa) {
+        return ipa;
+      }
+      if (stem.length >= 2) {
+        const gradated = applyFiGradation(stem);
+        if (gradated !== stem) {
+          const gradIpa = tryLookup(dict, ...replacements.map((r) => gradated + r));
+          if (gradIpa) {
+            return gradIpa;
+          }
+        }
+        const strengthened = applyFiStrengthening(stem);
+        if (strengthened !== stem) {
+          const strIpa = tryLookup(dict, ...replacements.map((r) => strengthened + r));
+          if (strIpa) {
+            return strIpa;
+          }
+        }
+      }
+    }
+  }
+  for (const poss of ["ni", "si", "nsa", "ns\xE4", "mme", "nne"]) {
+    if (word.endsWith(poss) && word.length > poss.length + 2) {
+      const inner = word.slice(0, -poss.length);
+      if (dict[inner]) {
+        return dict[inner];
+      }
+      for (const [suffix, replacements] of FI_SUFFIXES) {
+        if (inner.length > suffix.length + 1 && inner.endsWith(suffix)) {
+          const stem = inner.slice(0, -suffix.length);
+          const ipa = tryLookup(dict, ...replacements.map((r) => stem + r));
+          if (ipa) {
+            return ipa;
+          }
+        }
+      }
+    }
+  }
+  return void 0;
+}
+function modernizeNb(word) {
+  const variants = [];
+  if (word.includes("aa")) {
+    variants.push(word.replaceAll("aa", "\xE5"));
+  }
+  if (word.includes("Aa")) {
+    variants.push(word.replaceAll("Aa", "\xC5"));
+  }
+  if (word === "af") {
+    variants.push("av");
+  }
+  if (word === "efter") {
+    variants.push("etter");
+  }
+  if (word === "imod") {
+    variants.push("imot");
+  }
+  return variants;
+}
+var NB_SUFFIXES = [
+  // 4+ char
+  ["erne", ["", "e"]],
+  ["enes", ["", "e"]],
+  ["ande", ["", "e"]],
+  ["ende", ["", "e"]],
+  ["else", ["", "e"]],
+  // Definite plural
+  ["ene", ["", "e"]],
+  ["ane", ["", "e"]],
+  // 2 char definite singular
+  ["en", ["", "e"]],
+  ["et", ["", "e"]],
+  // Past tense / participle
+  ["te", ["", "e"]],
+  ["de", ["", "e"]],
+  ["dde", [""]],
+  // Indefinite plural
+  ["er", ["", "e"]],
+  // Comparative / superlative
+  ["ere", [""]],
+  ["est", [""]],
+  ["este", [""]],
+  // Present tense
+  ["ar", ["", "e"]],
+  // General
+  ["t", ["", "e"]],
+  ["a", ["", "e"]],
+  ["s", [""]],
+  ["e", [""]],
+  ["n", ["", "e"]],
+  ["r", ["", "e"]]
+];
+function lemmatizeNb(dict, word) {
+  const modern = modernizeNb(word);
+  for (const m of modern) {
+    if (dict[m]) {
+      return dict[m];
+    }
+    const lower = m.toLowerCase();
+    if (dict[lower]) {
+      return dict[lower];
+    }
+  }
+  for (const [suffix, replacements] of NB_SUFFIXES) {
+    if (word.length > suffix.length + 1 && word.endsWith(suffix)) {
+      const stem = word.slice(0, -suffix.length);
+      const ipa = tryLookup(dict, ...replacements.map((r) => stem + r));
+      if (ipa) {
+        return ipa;
+      }
+    }
+  }
+  for (const [suffix, replacements] of NB_SUFFIXES) {
+    if (word.length > suffix.length + 1 && word.endsWith(suffix)) {
+      const stem = word.slice(0, -suffix.length);
+      for (const r of replacements) {
+        const candidate = stem + r;
+        const modernized = modernizeNb(candidate);
+        for (const m of modernized) {
+          if (dict[m]) {
+            return dict[m];
+          }
+        }
+      }
+    }
+  }
+  return void 0;
+}
+var MA_SUFFIXES = ["nya", "mu", "ku", "kan", "an", "lah", "kah", "i"];
+var MA_PREFIXES = [
+  ["memper", [""]],
+  ["member", [""]],
+  ["menge", [""]],
+  ["meny", ["s", "c"]],
+  ["meng", ["k", "g", "h", ""]],
+  ["mem", ["p", "b", "f", ""]],
+  ["men", ["t", "d", "c", "j", ""]],
+  ["me", [""]],
+  ["diper", [""]],
+  ["ber", [""]],
+  ["per", [""]],
+  ["ter", [""]],
+  ["di", [""]],
+  ["ke", [""]],
+  ["se", [""]],
+  ["ku", [""]]
+];
+function lemmatizeMa(dict, word) {
+  for (const suffix of MA_SUFFIXES) {
+    if (word.length > suffix.length + 2 && word.endsWith(suffix)) {
+      const stem = word.slice(0, -suffix.length);
+      if (dict[stem]) {
+        return dict[stem];
+      }
+      const fromPrefix = tryMaPrefixStrip(dict, stem);
+      if (fromPrefix) {
+        return fromPrefix;
+      }
+    }
+  }
+  return tryMaPrefixStrip(dict, word);
+}
+function tryMaPrefixStrip(dict, word) {
+  for (const [prefix, restorations] of MA_PREFIXES) {
+    if (word.length > prefix.length + 1 && word.startsWith(prefix)) {
+      const remainder = word.slice(prefix.length);
+      if (dict[remainder]) {
+        return dict[remainder];
+      }
+      for (const consonant of restorations) {
+        if (consonant) {
+          const restored = consonant + remainder;
+          if (dict[restored]) {
+            return dict[restored];
+          }
+        }
+      }
+    }
+  }
+  return void 0;
+}
+var ZWNJ = "\u200C";
+function lemmatizeFa(dict, word) {
+  if (word.includes(ZWNJ)) {
+    const parts = word.split(ZWNJ);
+    for (const part of parts) {
+      if (dict[part]) {
+        return dict[part];
+      }
+    }
+    if (parts.length === 2 && (parts[0] === "\u0645\u06CC" || parts[0] === "\u0646\u0645\u06CC")) {
+      const verb = parts[1];
+      if (dict[verb]) {
+        return dict[verb];
+      }
+      for (const ending of ["\u0646\u062F", "\u0645", "\u06CC", "\u062F", "\u06CC\u0645", "\u06CC\u062F"]) {
+        if (verb.endsWith(ending) && verb.length > ending.length) {
+          const stem = verb.slice(0, -ending.length);
+          if (dict[stem]) {
+            return dict[stem];
+          }
+        }
+      }
+    }
+    const joined = parts.join("");
+    if (dict[joined]) {
+      return dict[joined];
+    }
+  }
+  for (const suffix of [
+    "\u0647\u0627\u06CC\u06CC",
+    "\u0647\u0627\u06CC",
+    "\u0647\u0627",
+    "\u0627\u06CC",
+    "\u0627\u0646",
+    "\u0627\u062A",
+    "\u06CC\u0646",
+    "\u062A\u0631",
+    "\u062A\u0631\u06CC\u0646",
+    "\u0634",
+    "\u0645",
+    "\u062A"
+  ]) {
+    if (word.endsWith(suffix) && word.length > suffix.length + 1) {
+      const stem = word.slice(0, -suffix.length);
+      if (dict[stem]) {
+        return dict[stem];
+      }
+    }
+  }
+  return void 0;
+}
+
+// src/overrides/ar.ts
+var ar = {
+  \u0623\u062D\u062F\u0627: "/\u0294a\u02C8\u0127adan/",
+  // anyone
+  \u0623\u062D\u062F\u0647\u0645\u0627: "/\u0294a\u0127adu\u02C8huma\u02D0/",
+  // one of them two
+  \u0623\u062D\u0631\u0627\u0631\u0627: "/\u0294a\u0127\u02C8ra\u02D0ran/",
+  \u0623\u0633\u062E\u064A\u0644\u064A\u0648\u0633: "/\u0294asxi\u02D0\u02C8li\u02D0ju\u02D0s/",
+  // Aeschylus
+  \u0623\u0645\u0647\u0627\u062A: "/um\u02D0a\u02C8ha\u02D0t/",
+  \u0623\u0646\u0633\u064A\u062A: "/\u0294un\u02C8si\u02D0t/",
+  \u0623\u0646\u0639\u0645\u062A: "/\u0294an\u02C8\u0295amta/",
+  \u0623\u0648\u0631\u0641\u0644\u064A\u0633: "/\u0294ur\u02C8fali\u02D0s/",
+  // Orphalese (Gibran)
+  \u0623\u064A\u062F\u064A\u0647\u0645: "/\u0294aj\u02C8di\u02D0him/",
+  \u0627\u0633\u062A\u064A\u0642\u0638\u062A: "/istaj\u02C8qa\xF0\u02E4tu/",
+  \u0627\u0633\u0645\u0647: "/\u02C8ismuhu/",
+  // his name
+  \u0627\u0639\u062A\u0632\u0627\u0644\u0647: "/i\u0295ti\u02C8za\u02D0lihi/",
+  // his seclusion
+  \u0627\u0644\u0623\u0642\u0627\u0648\u0627\u0644: "/al\u0294aqa\u02D0\u02C8wi\u02D0l/",
+  // the sayings/talk
+  \u0627\u0644\u0623\u0644\u064A\u0641\u0629: "/al\u0294a\u02C8li\u02D0fah/",
+  \u0627\u0644\u0623\u0648\u0644\u0649: "/al\u02C8\u0294u\u02D0la\u02D0/",
+  // the first (fem.)
+  \u0627\u0644\u062D\u064A\u0627\u0629: "/al\u0127a\u02C8ja\u02D0h/",
+  \u0627\u0644\u062E\u0644\u064A\u0641\u0629: "/alxa\u02C8li\u02D0fah/",
+  \u0627\u0644\u062E\u0644\u064A\u0642\u0629: "/alxa\u02C8li\u02D0qah/",
+  // creation/nature
+  \u0627\u0644\u0630\u0643\u0631\u064A\u0627\u062A: "/a\xF0\u02D0ikra\u02C8ja\u02D0t/",
+  \u0627\u0644\u0633\u0645\u0627\u0648\u0627\u062A: "/as\u02D0ama\u02D0\u02C8wa\u02D0t/",
+  \u0627\u0644\u0636\u0627\u0644\u064A\u0646: "/ad\u02E4\u02D0a\u02D0\u02C8li\u02D0n/",
+  \u0627\u0644\u0639\u0627\u0644\u0645\u064A\u0646: "/al\u0295a\u02D0la\u02C8mi\u02D0n/",
+  \u0627\u0644\u0639\u0634\u0631\u0629: "/al\u02C8\u0295a\u0283ara/",
+  // the ten
+  \u0627\u0644\u063A\u0631\u0628\u0629: "/al\u02C8\u0263urbah/",
+  \u0627\u0644\u063A\u0632\u0627\u0629: "/al\u0263u\u02C8za\u02D0h/",
+  \u0627\u0644\u0643\u0631\u0627\u0645\u0629: "/alka\u02C8ra\u02D0mah/",
+  \u0627\u0644\u0644\u0648\u0649: "/al\u02C8liwa\u02D0/",
+  \u0627\u0644\u0645\u0635\u0637\u0641\u0649: "/almus\u02C8t\u02E4afa\u02D0/",
+  \u0627\u0644\u0646\u062F\u0648\u0627\u062A: "/annada\u02C8wa\u02D0t/",
+  // the seminars/forums
+  \u0627\u0645\u0631\u0623\u0629: "/im\u02C8ra\u0294ah/",
+  \u0627\u0647\u062F\u0646\u0627: "/ih\u02C8dina\u02D0/",
+  \u0628\u0623\u0648\u0642\u0627\u0641\u0647: "/bi\u0294aw\u02C8qa\u02D0fihi/",
+  // with his endowments
+  \u0628\u0625\u0630\u0646\u0647: "/bi\u0294i\xF0\u02C8nihi/",
+  \u0628\u0628\u0631\u0642\u0629: "/bi\u02C8barqah/",
+  // in Barqa (place)
+  \u0628\u062C\u0632\u0627\u0626\u0631: "/bid\u0292a\u02C8za\u02D0\u0294ir/",
+  // with islands
+  \u0628\u062F\u0627\u064A\u0629: "/bi\u02C8da\u02D0jah/",
+  \u0628\u0630\u064A: "/bi\u02C8\xF0i\u02D0/",
+  // of importance
+  \u0628\u0634\u064A\u0621: "/bi\u02C8\u0283aj\u0294/",
+  \u0628\u0639\u0636\u0627: "/\u02C8ba\u0295d\u02E4an/",
+  \u0628\u0639\u0636\u0647\u0645: "/ba\u0295\u02C8d\u02E4uhum/",
+  \u0628\u0639\u064A\u0646\u0647: "/bi\u02C8\u0295ajnihi/",
+  \u0628\u0639\u064A\u0646\u064A\u0647\u0627: "/bi\u0295aj\u02C8najha\u02D0/",
+  // Egyptian National Anthem — بلادي بلادي
+  \u0628\u0644\u0627\u062F\u064A: "/bi\u02C8la\u02D0di\u02D0/",
+  // my country
+  \u0628\u064A\u062F\u0628\u0627: "/bajda\u02C8ba\u02D0/",
+  // Bidpai
+  \u062A\u0623\u062E\u0630\u0647: "/ta\u0294\u02C8xu\xF0uhu/",
+  \u062A\u0633\u0642\u0646\u064A: "/tas\u02C8qini\u02D0/",
+  \u062A\u0639\u0644\u0645\u062A: "/ta\u0295al\u02D0\u02C8amtu/",
+  \u062A\u0642\u0641\u0646: "/taqif\u02C8na/",
+  \u062A\u0642\u0644\u0628\u062A: "/taqal\u02D0a\u02C8bat/",
+  // it turned/fluctuated
+  \u062A\u0643\u0644\u0645\u064A: "/takal\u02D0a\u02C8mi\u02D0/",
+  // speak! (fem.)
+  \u062B\u0647\u0645\u062F: "/\u02C8\u03B8ahmad/",
+  // Thahmad (place name)
+  \u062C\u0644\u0633\u062A: "/d\u0292a\u02C8lasat/",
+  \u062D\u0627\u0631\u062A\u0646\u0627: "/\u0127a\u02D0\u02C8ratna\u02D0/",
+  \u062D\u0641\u0638\u0647\u0645\u0627: "/\u0127if\u02C8\xF0\u02E4ahuma\u02D0/",
+  \u062D\u0643\u0627\u064A\u0629: "/\u0127i\u02C8ka\u02D0jah/",
+  \u062D\u064A\u0627\u062A\u0647: "/\u0127a\u02C8ja\u02D0tahu/",
+  \u062E\u0641\u064A\u0641\u0627: "/xa\u02C8fi\u02D0fan/",
+  \u062E\u0644\u0627\u0644\u0647\u0627: "/xi\u02C8la\u02D0laha\u02D0/",
+  \u062E\u0644\u0641\u0647\u0645: "/xal\u02C8fahum/",
+  \u062E\u0645\u0631\u0627: "/\u02C8xamran/",
+  \u062F\u0628\u0634\u0644\u064A\u0645: "/dab\u0283a\u02C8li\u02D0m/",
+  // King Dabshelim
+  \u0631\u0627\u0626\u062D\u0629: "/\u02C8ra\u02D0\u0294i\u0127ah/",
+  \u0631\u0627\u0641\u0639\u0627: "/\u02C8ra\u02D0fi\u0295an/",
+  \u0631\u0633\u0645\u0647\u0627: "/ras\u02C8muha\u02D0/",
+  \u0631\u064A\u0627\u062D\u0643: "/ri\u02C8ja\u02D0\u0127aka/",
+  \u0632\u0639\u0645\u0648\u0627: "/za\u02C8\u0295amu\u02D0/",
+  \u0632\u0645\u0627\u0646\u0647: "/za\u02C8ma\u02D0nihi/",
+  \u0633\u0627\u062F\u062A\u064A: "/sa\u02D0\u02C8dati\u02D0/",
+  \u0633\u0627\u0633\u0627\u0646: "/sa\u02D0\u02C8sa\u02D0n/",
+  // Sasanian dynasty
+  \u0633\u0628\u0639\u0629: "/\u02C8sab\u0295ah/",
+  \u0633\u062A\u062D\u0645\u0644\u0647: "/sata\u0127\u02C8miluhu/",
+  \u0633\u0641\u064A\u0646\u062A\u0647: "/sa\u02C8fi\u02D0natuhu/",
+  \u0633\u0643\u0631\u0629: "/\u02C8sakrah/",
+  \u0633\u0646\u0629: "/\u02C8sanah/",
+  \u0633\u0646\u064A\u0646: "/si\u02C8ni\u02D0n/",
+  \u0634\u0647\u0631\u064A\u0627\u0631: "/\u0283ahri\u02C8ja\u02D0r/",
+  // King Shahryar (1001 Nights)
+  \u0634\u064A\u0626\u0627: "/\u02C8\u0283aj\u0294an/",
+  \u0635\u0628\u0627\u062D\u0627: "/s\u02E4a\u02C8ba\u02D0\u0127an/",
+  // morning
+  \u0635\u062D\u0628\u064A: "/\u02C8s\u02E4a\u0127bi\u02D0/",
+  // my companions
+  \u0635\u063A\u0627\u0631\u0647\u0627: "/s\u02E4i\u02C8\u0263a\u02D0ruha\u02D0/",
+  \u0637\u0627\u0644\u0639\u0627: "/\u02C8t\u02E4a\u02D0li\u0295an/",
+  \u0637\u0648\u064A\u0644\u0629: "/t\u02E4a\u02C8wi\u02D0lah/",
+  // Ibn Khaldun — المقدمة
+  \u0638\u0627\u0647\u0631\u0647: "/\xF0\u02E4a\u02D0\u02C8hirihi/",
+  // in its outward appearance
+  \u0639\u0627\u0645\u0627: "/\u02C8\u0295a\u02D0man/",
+  \u0639\u0628\u0644\u0629: "/\u02C8\u0295ablah/",
+  // Abla (name)
+  \u0639\u062C\u064A\u0628\u0629: "/\u0295a\u02C8d\u0292i\u02D0bah/",
+  \u0639\u062F\u062A: "/\u02C8\u0295udtu/",
+  \u0639\u0631\u0641\u062A: "/\u0295a\u02C8rafta/",
+  // you knew
+  \u0639\u0635\u0631\u0647: "/\u02C8\u0295as\u02E4rihi/",
+  \u0639\u0644\u0645\u0647: "/\u02C8\u0295ilmihi/",
+  \u0639\u0644\u064A\u0643: "/\u0295a\u02C8lajka/",
+  \u0639\u0644\u064A\u0647\u0645: "/\u0295a\u02C8lajhim/",
+  \u0639\u0646\u062F\u0647: "/\u02C8\u0295indahu/",
+  \u0639\u0646\u0647\u0627: "/\u0295an\u02C8ha\u02D0/",
+  // about it
+  \u063A\u0627\u064A\u062A\u064A: "/\u0263a\u02D0\u02C8jati\u02D0/",
+  // my goal
+  \u063A\u064A\u0628\u0629: "/\u02C8\u0263ajbah/",
+  \u0641\u0625\u0646: "/fa\u02C8\u0294in/",
+  \u0641\u0627\u0633\u0642\u0646\u064A: "/fas\u02C8qini\u02D0/",
+  \u0641\u0627\u063A\u062A\u0646\u0645\u0647\u0627: "/fa\u0263\u02C8tanimha\u02D0/",
+  \u0641\u0627\u0644\u0645\u0642\u0631\u0627\u0629: "/fal\u02C8miqra\u02D0h/",
+  \u0641\u062C\u0631\u0627: "/\u02C8fad\u0292ran/",
+  \u0641\u062D\u0648\u0645\u0644: "/fa\u02C8\u0127awmal/",
+  \u0641\u0645\u0627: "/fa\u02C8ma\u02D0/",
+  \u0641\u064A\u0647: "/fi\u02D0hi/",
+  \u0641\u064A\u0647\u0627: "/\u02C8fi\u02D0ha\u02D0/",
+  \u0642\u0627\u0644\u062A: "/\u02C8qa\u02D0lat/",
+  \u0642\u0635\u0629: "/\u02C8qis\u02E4\u02D0ah/",
+  \u0642\u0644\u062A: "/\u02C8qultu/",
+  // I said
+  \u0643\u062A\u0627\u0628\u0627\u062A: "/kita\u02D0\u02C8ba\u02D0t/",
+  \u0643\u0631\u0633\u064A\u0647: "/kur\u02C8sij\u02D0uhu/",
+  \u0644\u062E\u0648\u0644\u0629: "/lixaw\u02C8latah/",
+  // for Khawla (name)
+  // Tarafa ibn al-Abd — المعلقة
+  \u0644\u0643\u0650: "/\u02C8laki/",
+  // to you (fem.)
+  \u0644\u0646\u064A\u0644\u0643: "/li\u02C8najlika/",
+  // for your Nile
+  \u0644\u0648\u0647\u0644\u0629: "/li\u02C8wahlah/",
+  \u0645\u0628\u0646\u064A\u0629: "/mab\u02C8nij\u02D0ah/",
+  // Antara ibn Shaddad — المعلقة
+  \u0645\u062A\u0631\u062F\u0645: "/muta\u02C8rad\u02D0im/",
+  // worn out/in ruins
+  \u0645\u062A\u0642\u0637\u0639\u0627: "/mutaqat\u02E4\u02C8t\u02E4i\u0295an/",
+  \u0645\u062F\u064A\u0646\u0629: "/ma\u02C8di\u02D0nah/",
+  \u0645\u0637\u064A\u0647\u0645: "/mat\u02E4aj\u02D0a\u02C8hum/",
+  // their riding beasts
+  // Abu al-Ala al-Ma'arri — غير مجد في ملتي
+  \u0645\u0644\u062A\u064A: "/mil\u02D0a\u02C8ti\u02D0/",
+  // my religion/creed
+  \u0645\u0647\u064A\u0627\u0631: "/mah\u02C8ja\u02D0r/",
+  // Mahyar (Adonis)
+  \u0645\u064A\u0644\u0627\u062F\u0647: "/mi\u02D0\u02C8la\u02D0dihi/",
+  \u0646\u0633\u062C\u062A\u0647\u0627: "/nasa\u02C8d\u0292atha\u02D0/",
+  \u0646\u0648\u0645\u0627: "/\u02C8nawman/",
+  \u0647\u0627\u062F\u0645\u0627: "/\u02C8ha\u02D0diman/",
+  \u0647\u0648\u0630\u0627: "/ha\u02D0\u02C8\xF0a\u02D0/",
+  \u0648\u0623\u0639\u0648\u0627\u0646: "/wa\u0294a\u0295\u02C8wa\u02D0n/",
+  // and followers
+  \u0648\u0625\u0646\u0645\u0627: "/wa\u0294in\u02D0a\u02C8ma\u02D0/",
+  \u0648\u0625\u064A\u0627\u0643: "/wa\u0294ij\u02C8ja\u02D0ka/",
+  \u0648\u0627\u0633\u0644\u0645\u064A: "/wasla\u02C8mi\u02D0/",
+  // and farewell/be safe (fem.)
+  \u0648\u0627\u0639\u062A\u0642\u0627\u062F\u064A: "/wa\u0294i\u0295tiqa\u02D0\u02C8di\u02D0/",
+  // and my belief
+  \u0648\u0627\u0644\u0633\u0646\u0629: "/was\u02D0\u02C8anah/",
+  \u0648\u0627\u0644\u0642\u0627\u0644: "/wal\u02C8qa\u02D0l/",
+  // and the talk
+  \u0648\u0627\u0644\u0645\u062D\u0628\u0629: "/walma\u02C8\u0127ab\u02D0ah/",
+  \u0648\u0628\u0634\u0631\u0648\u0637\u0647: "/wabi\u0283u\u02C8ru\u02D0tihi/",
+  // and his conditions
+  \u0648\u062D\u0643\u0627\u064A\u0627\u062A: "/wa\u0127ika\u02D0\u02C8ja\u02D0t/",
+  \u0648\u0634\u0628\u064A\u0647: "/wa\u0283a\u02C8bi\u02D0h/",
+  // and similar to
+  \u0648\u0634\u0645\u0623\u0644: "/wa\u0283a\u02C8ma\u0294l/",
+  \u0648\u0636\u0639\u0647: "/\u02C8wad\u02E4\u0295ihi/",
+  \u0648\u0639\u0644\u0649: "/wa\u02C8\u0295ala\u02D0/",
+  // and upon
+  \u0648\u0639\u0644\u064A\u0647\u0645: "/wa\u0295a\u02C8lajhim/",
+  \u0648\u063A\u0627\u0628: "/wa\u02C8\u0263a\u02D0b/",
+  \u0648\u0641\u0624\u0627\u062F\u064A: "/wafu\u02C8\u0294a\u02D0di\u02D0/",
+  // and my heart
+  \u0648\u0641\u0637\u0646\u0629: "/wa\u02C8fit\u02E4nah/",
+  \u0648\u0642\u0648\u0641\u0627: "/wu\u02C8qu\u02D0fan/",
+  // standing
+  \u0648\u0645\u0627: "/wa\u02C8ma\u02D0/",
+  \u0648\u0645\u062A\u0633\u0627\u0648\u064A\u0646: "/wamutasa\u02D0\u02C8wi\u02D0n/",
+  \u0648\u0645\u0646\u0632\u0644: "/wa\u02C8manzil/",
+  \u0648\u0647\u0628\u0648\u0627: "/wu\u02C8hibu\u02D0/",
+  \u064A\u0624\u0648\u062F\u0647: "/ja\u0294u\u02D0\u02C8duhu/",
+  \u064A\u062A\u062E\u0637\u0649: "/jata\u02C8xat\u02E4\u02D0a\u02D0/",
+  \u064A\u062D\u064A\u0637\u0648\u0646: "/ju\u0127i\u02D0\u02C8t\u02E4u\u02D0n/",
+  \u064A\u0636\u0639\u0647: "/jad\u02E4a\u02C8\u0295uhu/",
+  \u064A\u0639\u0627\u0645\u0644\u0648\u0627: "/ju\u02C8\u0295a\u02D0milu\u02D0/",
+  \u064A\u0642\u0648\u0644\u0648\u0646: "/jaqu\u02D0\u02C8lu\u02D0n/",
+  // they say
+  \u064A\u0647\u062A\u0645\u0648\u0627: "/jahtam\u02C8mu\u02D0/"
+  // they cared
+};
+
+// src/overrides/de.ts
+var de = {
+  abendsonnenschein: "/\u02C8a\u02D0b\u0259nt\u02CCz\u0254n\u0259n\u0283a\u026A\u032Fn/",
+  // Foreign city names in German texts
+  Aires: "/\u02C8a\u026A\u032F\u0281\u025Bs/",
+  andrer: "/\u02C8and\u0281\u0250/",
+  balde: "/\u02C8bald\u0259/",
+  bewu\u00DFtsein: "/b\u0259\u02C8v\u028Ast\u02CCza\u026A\u032Fn/",
+  bl\u00FCtenschimmer: "/\u02C8bly\u02D0t\u0259n\u02CC\u0283\u026Am\u0250/",
+  brahmanensohn: "/\u02C8b\u0281a\u02D0ma\u02D0n\u0259n\u02CCzo\u02D0n/",
+  Buenos: "/\u02C8bue\u02D0n\u0254s/",
+  "c\u2019est": "/s\u025B/",
+  // French in Mann's Buddenbrooks
+  ch\u00E8re: "/\u0283\u025B\u02D0\u0281/",
+  // French in Mann's Buddenbrooks
+  da\u00DF: "/das/",
+  "davos-platz": "/\u02C8da\u02D0v\u0254s\u02CCplats/",
+  // Swiss resort area
+  demoiselle: "/d\u0259mwa\u02C8z\u025Bl/",
+  // French in Mann's Buddenbrooks
+  d\u00FCwel: "/\u02C8dy\u02D0v\u0259l/",
+  // Low German: devil
+  fa\u00DFt: "/fast/",
+  feuertrunken: "/\u02C8f\u0254\u026A\u032F\u0250\u02CCt\u0281\u028A\u014Bk\u0259n/",
+  flu\u00DFufers: "/\u02C8fl\u028As\u02CC\u0294u\u02D0f\u0250s/",
+  fr\u00FChlingsnachmittag: "/\u02C8f\u0281y\u02D0l\u026A\u014Bs\u02CCna\u02D0xm\u026Ata\u02D0k/",
+  gefahrdrohende: "/\u0261\u0259\u02C8fa\u02D0\u0250\u032F\u02CCd\u0281o\u02D0\u0259nd\u0259/",
+  gek\u00FC\u00DFt: "/\u0261\u0259\u02C8k\u028Fst/",
+  g\u00F6tterfunken: "/\u02C8\u0261\u0153t\u0250\u02CCf\u028A\u014Bk\u0259n/",
+  govinda: "/\u0261o\u02C8v\u026Anda/",
+  // Sanskrit name
+  graub\u00FCndischen: "/\u0261\u0281a\u028A\u032F\u02C8b\u028Fnd\u026A\u0283\u0259n/",
+  guizot: "/\u0261i\u02C8zo\u02D0/",
+  // French name
+  h\u00E4tt: "/h\u025Bt/",
+  // contraction of hätte
+  "ich\u2019s": "/\u026A\xE7s/",
+  k\u00FC\u00DFnacht: "/\u02C8k\u028Fsnaxt/",
+  macheath: "/m\u0259\u02C8ki\u02D0\u03B8/",
+  // English name
+  metternich: "/\u02C8m\u025Bt\u0250n\u026A\xE7/",
+  m\u00FC\u00DFt: "/m\u028Fst/",
+  mu\u00DFte: "/\u02C8m\u028Ast\u0259/",
+  New: "/nju\u02D0/",
+  ook: "/o\u02D0k/",
+  // Low German: also
+  "prinz-regentenstra\xDFe": "/\u02C8p\u0281\u026Ants\u0281e\u02CC\u0261\u025Bnt\u0259n\u02CC\u0283t\u0281a\u02D0s\u0259/",
+  question: "/k\u025Bs\u02C8tj\u0254\u0303/",
+  // French in Mann's Buddenbrooks
+  salwaldes: "/\u02C8zal\u02CCvald\u0259s/",
+  // Sal forest
+  samsa: "/\u02C8zamza/",
+  // Kafka character
+  schriee: "/\u02C8\u0283\u0281i\u02D0\u0259/",
+  siddhartha: "/z\u026A\u02C8da\u02D0\u0281ta/",
+  sternklar: "/\u02C8\u0283t\u025B\u0281nkla\u02D0\u0250\u032F/",
+  tr\u00E8s: "/t\u0281\u025B/",
+  // French in Mann's Buddenbrooks
+  verliess: "/f\u025B\u0250\u032F\u02C8li\u02D0s/",
+  York: "/j\u0254\u0281k/",
+  zarathustra: "/tsa\u0281a\u02C8t\u028Ast\u0281a/"
+};
+
+// src/overrides/eo.ts
+var eo = {
+  // G2P handles all Esperanto words with correct penultimate stress.
+  // Overrides here are only needed for words where the IPA dictionary
+  // has an incorrect entry that needs correction.
+};
+
+// src/overrides/es.ts
+var es = {
+  arts: "/a\u0281/",
+  // French "Arts" in Cortázar (Pont des Arts)
+  aureliano: "/aw\u027Ee\u02C8ljano/",
+  beatriz: "/bea\u02C8t\u027Eis/",
+  buend\u00EDa: "/bwen\u02C8dia/",
+  ca\u00F1abrava: "/ka\u0272a\u02C8\u03B2\u027Ea\u03B2a/",
+  conti: "/\u02C8konti/",
+  // Italian name in Cortázar
+  fierro: "/\u02C8fje\u027Eo/",
+  macondo: "/ma\u02C8kondo/",
+  pont: "/p\u0254\u0303/",
+  // French in Cortázar
+  porfirio: "/po\u027E\u02C8fi\u027Ejo/",
+  quai: "/ke/",
+  // French in Cortázar
+  qued\u00F3se: "/ke\u02C8\xF0ose/",
+  rue: "/\u0281y/",
+  // French in Cortázar
+  seine: "/s\u025Bn/",
+  // French river in Cortázar
+  s\u00E9pase: "/\u02C8sepase/",
+  urbino: "/u\u027E\u02C8\u03B2ino/",
+  vacilante: "/basi\u02C8lante/",
+  viterbo: "/bi\u02C8te\u027E\u03B2o/"
+};
+
+// src/overrides/fa.ts
+var fa = {
+  \u0622\u0628\u0650: "/\u0252\u02D0be/",
+  \u0622\u0631\u0627\u06CC: "/\u0252\u02D0\u027E\u0252\u02D0j/",
+  \u0622\u0641\u0631\u06CC\u0646: "/\u0252\u02D0f\xE6\u027Ei\u02D0n/",
+  // creator
+  \u0622\u0645\u0648\u06CC: "/\u0252\u02D0mu\u02D0j/",
+  \u0622\u0648\u0627\u0631\u0647: "/\u0252\u02D0v\u0252\u02D0\u027Ee/",
+  // wandering, exiled
+  \u0622\u0648\u0627\u0632\u0647: "/\u0252\u02D0v\u0252\u02D0ze/",
+  // fame, renown
+  \u0622\u06CC\u064E\u062F: "/\u0252\u02D0j\xE6d/",
+  // --- Forough Farrokhzad, Another Birth ---
+  \u0622\u06CC\u0647: "/\u0252\u02D0je/",
+  // verse (of scripture)
+  \u0627\u0628\u062F\u06CC: "/\xE6b\xE6di\u02D0/",
+  // eternal
+  \u0627\u0628\u0648\u0627\u0644\u0641\u0636\u0644: "/\xE6bolf\xE6zl/",
+  \u0627\u062C\u062A\u0645\u0627\u0639\u06CC: "/ed\u0361\u0292tem\u0252\u02D0\u0294i\u02D0/",
+  \u0627\u062F\u0631: "/\xE6de\u027E/",
+  // pour! (Arabic imperative)
+  \u0627\u0633\u0627\u0633: "/\xE6s\u0252\u02D0s/",
+  // --- Iranian Constitution ---
+  \u0627\u0633\u0627\u0633\u06CC: "/\xE6s\u0252\u02D0si\u02D0/",
+  \u0627\u0635\u0648\u0644: "/osu\u02D0l/",
+  \u0627\u0641\u062A\u0627\u062F: "/oft\u0252\u02D0d/",
+  // fell, befell
+  // --- UDHR Article 1 ---
+  \u0627\u0641\u0631\u0627\u062F: "/\xE6f\u027E\u0252\u02D0d/",
+  // individuals
+  \u0627\u0642\u062A\u0635\u0627\u062F\u06CC: "/e\u0262tes\u0252\u02D0di\u02D0/",
+  // --- Hafez, Divan (Ghazal 1) ---
+  \u0627\u0644\u0627: "/\xE6l\u0252\u02D0/",
+  // O! behold
+  \u0627\u0644\u0633\u0627\u0642\u06CC: "/\xE6ss\u0252\u02D0\u0262i\u02D0/",
+  // the cupbearer (Arabic definite)
+  \u0627\u0644\u0644\u0651\u0647: "/\xE6ll\u0252\u02D0h/",
+  \u0627\u0646\u062F\u0631\u0634: "/\xE6nd\xE6\u027E\xE6\u0283/",
+  // within it
+  \u0627\u0646\u0632\u0648\u0627: "/enzev\u0252\u02D0/",
+  // seclusion
+  \u0627\u0646\u0639\u06A9\u0627\u0633: "/en\u0294ek\u0252\u02D0s/",
+  // --- Sepehri ---
+  \u0627\u0647\u0644: "/\xE6hl/",
+  \u0627\u06CC: "/ej/",
+  // O! (vocative)
+  \u0627\u06CC\u0647\u0627: "/\xE6jjoh\u0252\u02D0/",
+  // O! (Arabic vocative)
+  \u0628\u0627\u062F\u0647\u0627\u06CC: "/b\u0252\u02D0dh\u0252\u02D0je/",
+  // winds (with ezafeh)
+  "\u0628\u0628\u0631\u06CC\u062F\u0647\u200C\u0627\u0646\u062F": "/bob\u027Ei\u02D0de\xE6nd/",
+  // they have cut
+  // --- Baba Taher ---
+  \u0628\u0628\u0646\u062F\u0645: "/beb\xE6nd\xE6m/",
+  \u0628\u062A\u0627\u0628\u062F: "/bet\u0252\u02D0b\xE6d/",
+  // will shine
+  \u0628\u062C\u0627\u0646: "/bed\u0361\u0292\u0252\u02D0n/",
+  // --- Attar, Conference of the Birds ---
+  \u0628\u062E\u0634\u06CC\u062F: "/b\xE6x\u0283i\u02D0d/",
+  \u0628\u062F\u0627\u0631: "/bed\u0252\u02D0\u027E/",
+  \u0628\u062F\u0627\u0634\u062A: "/bed\u0252\u02D0\u0283t/",
+  \u0628\u0631\u062A\u0631: "/b\xE6\u027Et\xE6\u027E/",
+  // higher, superior
+  \u0628\u0631\u062F: "/bo\u027Ed/",
+  // will carry (subjunctive)
+  \u0628\u0633\u062A\u0646\u062F: "/b\xE6st\xE6nd/",
+  \u0628\u0634\u0627\u06CC\u062F: "/be\u0283\u0252\u02D0j\xE6d/",
+  \u0628\u0634\u0631: "/b\xE6\u0283\xE6\u027E/",
+  // humankind
+  // --- Rumi, Masnavi (Song of the Reed) ---
+  \u0628\u0634\u0646\u0648: "/be\u0283no/",
+  // listen!
+  \u0628\u0634\u0648\u06CC\u0645: "/be\u0283u\u02D0j\xE6m/",
+  \u0628\u06AF\u0631\u062F\u0645: "/be\u0261\xE6\u027Ed\xE6m/",
+  \u0628\u06AF\u0634\u0627\u06CC: "/be\u0261o\u0283\u0252\u02D0j/",
+  \u0628\u06AF\u0648\u06CC\u0645: "/be\u0261u\u02D0j\xE6m/",
+  // let me say
+  \u0628\u0646\u0627\u0632\u0645: "/ben\u0252\u02D0z\xE6m/",
+  // --- Saadi, Bustan ---
+  \u0628\u0646\u0627\u0645: "/ben\u0252\u02D0m/",
+  // in the name of
+  \u0628\u0646\u0634\u0633\u062A\u0647: "/bene\u0283\xE6ste/",
+  // has sat
+  \u0628\u0646\u0645\u0627\u06CC: "/ben\xE6m\u0252\u02D0j/",
+  // --- Rudaki ---
+  \u0628\u0648\u06CC\u0650: "/bu\u02D0je/",
+  \u0628\u06CC\u0647\u0642\u06CC: "/bejh\xE6\u0262i\u02D0/",
+  \u067E\u064E\u0631\u0646\u06CC\u0627\u0646: "/p\xE6\u027Eni\u02D0\u0252\u02D0n/",
+  // --- Sanai ---
+  \u067E\u0631\u0648\u0631: "/p\xE6\u027Ev\xE6\u027E/",
+  \u067E\u0633\u062A\u06CC: "/p\xE6sti\u02D0/",
+  "\u067E\u0648\u0632\u0634\u200C\u067E\u0630\u06CC\u0631": "/pu\u02D0ze\u0283p\xE6zi\u02D0\u027E/",
+  // excuse-accepting
+  \u062A\u0627\u0631\u06CC\u06A9\u06CC\u0633\u062A: "/t\u0252\u02D0\u027Ei\u02D0ki\u02D0st/",
+  // is a darkness (contraction)
+  \u062A\u062D\u0642\u06CC\u0642: "/t\xE6h\u0262i\u02D0\u0262/",
+  \u062A\u0648\u0641\u06CC\u0642: "/to\u028Afi\u02D0\u0262/",
+  "\u062C\u062F\u0627\u06CC\u06CC\u200C\u0647\u0627": "/d\u0361\u0292od\u0252\u02D0ji\u02D0h\u0252\u02D0/",
+  // separations
+  \u062C\u0644: "/d\u0361\u0292\xE6ll/",
+  // majesty
+  \u062C\u0648\u0627\u0628\u0650: "/d\u0361\u0292\xE6v\u0252\u02D0be/",
+  \u062C\u0648\u06CC\u0650: "/d\u0361\u0292u\u02D0je/",
+  \u062D\u0627\u0641\u0638: "/h\u0252\u02D0fez/",
+  \u062D\u0627\u0644: "/h\u0252\u02D0l/",
+  // state, condition
+  \u062D\u0627\u0644\u06CC: "/h\u0252\u02D0li\u02D0/",
+  // pleasant (archaic: now)
+  \u062D\u0642\u0648\u0642: "/ho\u0262u\u02D0\u0262/",
+  // rights
+  \u062D\u06A9\u06CC\u0645: "/h\xE6ki\u02D0m/",
+  // wise
+  \u062D\u06CC\u0627\u062A: "/h\xE6j\u0252\u02D0t/",
+  // life
+  \u062D\u06CC\u062B\u06CC\u062A: "/hejsi\u02D0j\xE6t/",
+  // dignity
+  // --- Nezami, Khosrow o Shirin ---
+  \u062E\u062F\u0627\u0648\u0646\u062F\u0627: "/xod\u0252\u02D0v\xE6nd\u0252\u02D0/",
+  // --- Saadi, Golestan ---
+  \u062E\u062F\u0627\u06CC: "/xod\u0252\u02D0j/",
+  // God (with ezafeh)
+  \u062E\u0631\u062F\u0628\u062E\u0634: "/xe\u027E\xE6db\xE6x\u0283/",
+  \u062E\u0637\u0627\u0628\u062E\u0634: "/x\xE6t\u0252\u02D0b\xE6x\u0283/",
+  // sin-forgiving
+  \u062E\u0650\u0646\u06AF\u0650: "/xen\u0261e/",
+  \u062E\u0648\u0634\u0627: "/xo\u0283\u0252\u02D0/",
+  \u062E\u0648\u0634\u062E\u0648\u0627\u0646: "/xo\u0283x\u0252\u02D0n/",
+  // sweet-singing
+  // --- Behbahani ---
+  \u062E\u0648\u06CC\u0634: "/xi\u02D0\u0283/",
+  \u062E\u06CC\u0632\u0631\u0627\u0646: "/xi\u02D0z\u027E\u0252\u02D0n/",
+  // bamboo
+  \u062F\u0627\u0631\u0627\u06CC: "/d\u0252\u02D0\u027E\u0252\u02D0je/",
+  // possessing
+  \u062F\u0627\u0646\u0646\u062F: "/d\u0252\u02D0n\xE6nd/",
+  // they know
+  \u062F\u0631\u0650: "/d\xE6\u027Ee/",
+  \u062F\u064F\u0631\u0634\u062A\u06CC\u200C\u0647\u0627\u06CC: "/do\u027Eo\u0283ti\u02D0h\u0252\u02D0je/",
+  \u062F\u0631\u06CC\u0686\u0647\u200C\u06CC: "/d\xE6\u027Ei\u02D0t\u0361\u0283eje/",
+  \u062F\u0633\u062A\u06AF\u06CC\u0631: "/d\xE6st\u0261i\u02D0\u027E/",
+  // helper
+  \u062F\u0644\u06CC: "/deli/",
+  \u062F\u0648\u0633\u062A\u0627\u0646\u06CC: "/du\u02D0st\u0252\u02D0ni\u02D0/",
+  \u062F\u06CC\u062F\u0627\u0631: "/di\u02D0d\u0252\u02D0\u027E/",
+  \u0630\u0648\u0642\u06CC: "/zo\u028A\u0262i\u02D0/",
+  \u0631\u0627: "/\u027E\u0252\u02D0/",
+  // object marker
+  \u0631\u0627\u0632\u0642: "/\u027E\u0252\u02D0ze\u0262/",
+  "\u0631\u0633\u062A\u0646\u200C\u0647\u0627\u06CC": "/\u027E\xE6st\xE6nh\u0252\u02D0je/",
+  // growings (with ezafeh)
+  \u0631\u0636\u06CC: "/\u027E\xE6zi\u02D0/",
+  \u0631\u0641\u062A\u0627\u0631: "/\u027E\xE6ft\u0252\u02D0\u027E/",
+  // behavior
+  \u0631\u0647: "/\u027E\xE6h/",
+  \u0631\u0647\u0650: "/\u027E\xE6he/",
+  \u0631\u0647\u0646\u0645\u0627\u06CC: "/\u027E\xE6hn\xE6m\u0252\u02D0j/",
+  // guide
+  \u0631\u0648\u062D\u06CC\u0647: "/\u027Eu\u02D0hi\u02D0je/",
+  // spirit, morale
+  \u0631\u0648\u0632\u06AF\u0627\u0631\u0645: "/\u027Eu\u02D0ze\u0261\u0252\u02D0\u027E\xE6m/",
+  "\u0631\u0648\u0632\u06CC\u200C\u062F\u0647": "/\u027Eu\u02D0zi\u02D0deh/",
+  // sustenance-giver
+  \u0631\u06CC\u06AF\u0650: "/\u027Ei\u02D0\u0261e/",
+  \u0632\u0627\u06CC\u06CC\u062F\u0647: "/z\u0252\u02D0ji\u02D0de/",
+  // born
+  \u0632\u0628\u0627\u0646\u06CC: "/z\xE6b\u0252\u02D0ni\u02D0/",
+  // --- Hedayat, The Blind Owl ---
+  "\u0632\u062E\u0645\u200C\u0647\u0627\u06CC\u06CC": "/z\xE6xmh\u0252\u02D0ji\u02D0/",
+  // wounds (indefinite)
+  \u0632\u06CC: "/zi\u02D0/",
+  \u0632\u06CC\u0652: "/zi\u02D0/",
+  "\u0633\u0627\u062D\u0644\u200C\u0647\u0627": "/s\u0252\u02D0helh\u0252\u02D0/",
+  // shores
+  \u0633\u0628\u06A9\u0628\u0627\u0631\u0627\u0646: "/s\xE6bokb\u0252\u02D0\u027E\u0252\u02D0n/",
+  // light-laden ones
+  \u0633\u0628\u06A9\u062A\u06AF\u06CC\u0646: "/sobokte\u0261i\u02D0n/",
+  \u0633\u062D\u0631\u06AF\u0627\u0647: "/s\xE6h\xE6\u027E\u0261\u0252\u02D0h/",
+  // dawn
+  // --- Nezami, Layli o Majnun ---
+  \u0633\u0631\u0622\u063A\u0627\u0632: "/s\xE6r\u0252\u02D0\u0261\u0252\u02D0z/",
+  \u0633\u0631\u0627\u06CC\u062F: "/so\u027E\u0252\u02D0j\xE6d/",
+  \u0633\u0644\u0627\u0645\u06CC: "/s\xE6l\u0252\u02D0mi\u02D0/",
+  // --- Khayyam, Rubaiyat ---
+  \u0633\u0648\u062F\u0627: "/so\u028Ad\u0252\u02D0/",
+  // passion, melancholy
+  \u0633\u06CC\u0627\u0633\u06CC: "/sij\u0252\u02D0si\u02D0/",
+  \u0634\u0627\u062F\u0645\u0627\u0646: "/\u0283\u0252\u02D0dm\u0252\u02D0n/",
+  \u0634\u0631\u062D: "/\u0283\xE6\u027Eh/",
+  // explanation
+  \u0634\u0631\u062D\u0647: "/\u0283\xE6\u027Ehe/",
+  // explanation (with ezafeh)
+  \u0634\u06A9\u0627\u06CC\u062A: "/\u0283ek\u0252\u02D0j\xE6t/",
+  // complaint
+  "\u0634\u06A9\u0641\u062A\u0646\u200C\u0647\u0627": "/\u0283ekoft\xE6nh\u0252\u02D0/",
+  // blossomings
+  \u0635\u0644\u062A\u0650: "/sele\u02D0te/",
+  \u0635\u0646\u0639: "/son\u0294/",
+  \u0636\u0648\u0627\u0628\u0637: "/z\xE6v\u0252\u02D0bet/",
+  \u0637\u0627\u0639\u062A\u0634: "/t\u0252\u02D0\u0294\xE6t\xE6\u0283/",
+  // his obedience
+  \u0639\u0632: "/\u0294\xE6zz/",
+  // glory
+  \u0639\u0645\u0631: "/om\u027E/",
+  \u0639\u0646\u0647: "/\xE6nho/",
+  \u063A\u0627\u0632\u06CC: "/\u0263\u0252\u02D0zi\u02D0/",
+  \u063A\u0627\u06CC\u062A: "/\u0263\u0252\u02D0j\xE6t/",
+  \u063A\u0632\u0644: "/\u0263\xE6z\xE6l/",
+  \u063A\u0632\u0646\u0648\u06CC: "/\u0263\xE6zn\xE6vi\u02D0/",
+  \u063A\u0632\u0646\u06CC: "/\u0263\xE6zni\u02D0/",
+  \u0641\u0631\u0627\u0642: "/fe\u027E\u0252\u02D0\u0262/",
+  // separation
+  \u0641\u0631\u062F: "/f\xE6\u027Ed/",
+  // alone, single
+  \u0641\u0631\u0647\u0646\u06AF\u06CC: "/f\xE6rh\xE6n\u0261i\u02D0/",
+  \u0641\u0631\u0648: "/fo\u027Eu\u02D0/",
+  // down
+  \u0641\u0644\u06A9: "/f\xE6l\xE6k/",
+  \u0642\u062F\u06A9: "/\u0262\xE6d\xE6k/",
+  \u0642\u0631\u0628\u062A: "/\u0262o\u027Eb\xE6t/",
+  // nearness (to God)
+  \u0642\u0635\u06CC\u062F\u0647\u200C\u0627\u06CC: "/\u0262\xE6si\u02D0de\xE6j/",
+  // --- Nima Yushij, The Phoenix ---
+  \u0642\u0642\u0646\u0648\u0633: "/\u0262o\u0262nu\u02D0s/",
+  // phoenix
+  \u06A9\u0622\u0641\u0631\u06CC\u0646\u062A: "/k\u0252\u02D0f\xE6\u027Ei\u02D0n\xE6t/",
+  \u06A9\u0627\u0631\u06AF\u0634\u0627\u06CC: "/k\u0252\u02D0\u027E\u0261o\u0283\u0252\u02D0j/",
+  \u06A9\u0627\u0633\u0627: "/k\xE6\u0294s\u0252\u02D0/",
+  // a cup (Arabic accusative, Persianized)
+  \u06A9\u0627\u0634\u0627\u0646\u0645: "/k\u0252\u02D0\u0283\u0252\u02D0n\xE6m/",
+  \u06A9\u0627\u0648: "/k\u0252\u02D0w/",
+  \u06A9\u0631\u06CC\u0645: "/k\xE6\u027Ei\u02D0m/",
+  // generous
+  // --- Ferdowsi, Shahnameh ---
+  \u06A9\u0632: "/k\xE6z/",
+  // contraction of که از (that from)
+  \u06A9\u0634\u06CC\u062F\u0645: "/ke\u0283i\u02D0d\xE6m/",
+  // I drew (a sigh)
+  \u06A9\u0644\u0627\u0645: "/k\xE6l\u0252\u02D0m/",
+  \u06A9\u0646\u0627\u0646: "/kon\u0252\u02D0n/",
+  // doing (present participle suffix)
+  \u06A9\u0646\u0645: "/kon\xE6m/",
+  \u06A9\u0646\u0646\u062F: "/kon\xE6nd/",
+  // they do (subjunctive)
+  \u06AF\u0631\u062F\u0627\u0628\u06CC: "/\u0261e\u027Ed\u0252\u02D0bi\u02D0/",
+  // a whirlpool (indefinite)
+  // --- Akhavan-Sales ---
+  \u06AF\u0641\u062A: "/\u0261oft/",
+  \u06AF\u064F\u0644: "/\u0261ol/",
+  \u06AF\u0648\u06CC\u062F: "/\u0261u\u02D0j\xE6d/",
+  // --- Beyhaqi ---
+  \u06AF\u0648\u06CC\u0646\u062F\u0647\u0654: "/\u0261u\u02D0j\xE6ndeje/",
+  \u0644\u063A\u0632\u0627\u0646: "/l\xE6\u0263z\u0252\u02D0n/",
+  \u0645\u0627\u062F\u0631\u06CC: "/m\u0252\u02D0d\xE6\u027Ei\u02D0/",
+  \u0645\u0627\u0646\u062F\u0647: "/m\u0252\u02D0nde/",
+  // remaining
+  \u0645\u0627\u0647\u062A\u0627\u0628: "/m\u0252\u02D0ht\u0252\u02D0b/",
+  // moonlight
+  \u0645\u0628\u06CC\u0646: "/mob\xE6jjen/",
+  \u0645\u062D\u0645\u0648\u062F: "/m\xE6hmu\u02D0d/",
+  \u0645\u062F\u0647: "/m\xE6deh/",
+  \u0645\u0632\u06CC\u062F: "/m\xE6zi\u02D0d/",
+  // increase
+  \u0645\u0641\u0631\u062D: "/mof\xE6\u027E\u027Eeh/",
+  // gladdening
+  \u0645\u0645\u062F: "/momedd/",
+  // sustainer
+  \u0645\u0648\u062C\u0628: "/mo\u028Ad\u0361\u0292eb/",
+  // cause of
+  \u0645\u0648\u0644\u06CC\u0627\u0646: "/mu\u02D0li\u02D0\u0252\u02D0n/",
+  \u0645\u0648\u0646\u0633: "/mu\u02D0nes/",
+  \u0645\u0648\u0647\u0628\u062A: "/mo\u028Aheb\xE6t/",
+  // gift, endowment
+  "\u0645\u06CC\u200C\u0622\u06CC\u062F": "/mi\u02D0\u0252\u02D0j\xE6d/",
+  // comes (up)
+  "\u0645\u06CC\u200C\u062A\u0631\u0627\u0634\u062F": "/mi\u02D0t\xE6\u027E\u0252\u02D0\u0283\xE6d/",
+  // scrapes, carves
+  "\u0645\u06CC\u200C\u062E\u0648\u0631\u062F": "/mi\u02D0xo\u027E\xE6d/",
+  // eats, erodes
+  "\u0645\u06CC\u200C\u0631\u0648\u062F": "/mi\u02D0\u027E\xE6v\xE6d/",
+  // goes (down)
+  "\u0645\u06CC\u200C\u0634\u0648\u0646\u062F": "/mi\u02D0\u0283\xE6v\xE6nd/",
+  // they become (passive)
+  "\u0645\u06CC\u200C\u06A9\u0646\u062F": "/mi\u02D0kon\xE6d/",
+  // does, is doing
+  \u0646\u0627\u067E\u0633\u0646\u062F\u200C\u0645: "/n\u0252\u02D0p\xE6s\xE6nd\xE6m/",
+  \u0646\u0627\u062E\u0648\u0628: "/n\u0252\u02D0xu\u02D0b/",
+  \u0646\u0627\u0635\u0631: "/n\u0252\u02D0se\u027E/",
+  "\u0646\u0627\u0644\u06CC\u062F\u0647\u200C\u0627\u0646\u062F": "/n\u0252\u02D0li\u02D0de\xE6nd/",
+  // they have lamented
+  \u0646\u0627\u0646\u06CC: "/n\u0252\u02D0ni\u02D0/",
+  \u0646\u0627\u0648\u0644\u0647\u0627: "/n\u0252\u02D0velh\u0252\u02D0/",
+  // pass it around
+  \u0646\u062A\u0648\u0627\u0646\u062F: "/n\xE6t\xE6v\u0252\u02D0n\xE6d/",
+  \u0646\u0634\u0627\u0637\u0650: "/ne\u0283\u0252\u02D0te/",
+  \u0646\u0638\u0627\u0645\u06CC: "/nez\u0252\u02D0mi\u02D0/",
+  \u0646\u0638\u0631\u0628\u0627\u0632\u06CC\u0627: "/n\xE6z\xE6\u027Eb\u0252\u02D0zi\u02D0\u0252\u02D0/",
+  \u0646\u0639\u0645\u062A: "/ne\u0294m\xE6t/",
+  // blessing
+  \u0646\u0641\u0633\u06CC: "/n\xE6f\xE6si\u02D0/",
+  // a breath (indefinite)
+  \u0646\u0641\u06CC\u0631\u0645: "/n\xE6fi\u02D0\u027E\xE6m/",
+  // my wailing
+  \u0646\u06AF\u0627\u0647\u0650: "/ne\u0261\u0252\u02D0he/",
+  \u0646\u06AF\u0630\u0631\u062F: "/n\xE6\u0261oz\xE6\u027E\xE6d/",
+  // does not pass
+  \u0646\u06AF\u0647: "/ne\u0261\xE6h/",
+  \u0646\u0645\u0648\u062F: "/nemu\u02D0d/",
+  // appeared, showed
+  "\u0646\u0645\u06CC\u200C\u0634\u0648\u062F": "/n\xE6mi\u02D0\u0283\xE6v\xE6d/",
+  // is not possible
+  // --- Shared classical forms ---
+  \u0646\u0647\u0627\u062F: "/neh\u0252\u02D0d/",
+  \u0646\u0647\u0627\u062F\u0647\u0627\u06CC: "/n\xE6h\u0252\u02D0dh\u0252\u02D0je/",
+  \u0646\u0648\u0634: "/nu\u02D0\u0283/",
+  // drink!
+  \u0646\u06CC\u0627\u0628\u062F: "/n\xE6j\u0252\u02D0b\xE6d/",
+  // will not find
+  \u0646\u06CC\u0627\u0631\u062F: "/n\xE6j\u0252\u02D0\u027E\xE6d/",
+  \u0646\u06CC\u0633\u062A\u0627\u0646: "/nejest\u0252\u02D0n/",
+  // reed bed
+  // --- Shamlou ---
+  \u0646\u06CC\u0633\u062A\u06CC: "/ni\u02D0sti\u02D0/",
+  \u0647\u0627\u06CC\u0644: "/h\u0252\u02D0jel/",
+  // terrifying
+  \u0647\u0633\u062A\u0646\u062F: "/h\xE6st\xE6nd/",
+  // they are
+  \u0647\u064E\u0645\u06CC: "/h\xE6mi\u02D0/",
+  \u0647\u0648\u0634\u06CC: "/hu\u02D0\u0283i\u02D0/",
+  \u0648\u062C\u062F\u0627\u0646: "/ved\u0361\u0292d\u0252\u02D0n/",
+  // conscience
+  \u0648\u062F\u06CC\u0639\u062A: "/v\xE6di\u02D0\xE6t/",
+  \u0648\u0632\u0634: "/v\xE6ze\u0283/",
+  // blowing
+  \u0648\u06CC: "/v\xE6j/",
+  \u06CC\u0627\u062F\u0650: "/j\u0252\u02D0de/",
+  \u06CC\u0627\u0631\u0650: "/j\u0252\u02D0\u027Ee/",
+  \u06CC\u0627\u0641\u062A: "/j\u0252\u02D0ft/",
+  \u06CC\u06A9\u062F\u06CC\u06AF\u0631: "/jekdi\u02D0\u0261\xE6\u027E/"
+  // one another
+};
+
+// src/overrides/fi.ts
+var fi = {
+  // G2P handles all Finnish words correctly. Overrides here are only
+  // needed for words where G2P produces an incorrect entry.
+};
+
+// src/overrides/fr.ts
+var fr = {
+  conflans: "/k\u0254\u0303fl\u0251\u0303/",
+  // place name
+  est: "/\u025B/",
+  // verb "is" — st is silent (dict has /ɛst/)
+  jolies: "/\u0292\u0254li/",
+  luit: "/l\u0265i/",
+  marchiennes: "/ma\u0281\u0283j\u025Bn/",
+  // place name (Zola)
+  "m\xE9taphysico-th\xE9ologo-cosmolonigologie": "/metafizikoteol\u0254\u0261\u0254k\u0254sm\u0254l\u0254ni\u0261\u0254l\u0254\u0292i/",
+  // Voltaire
+  montsou: "/m\u0254\u0303su/",
+  // fictional town (Zola)
+  morgion: "/m\u0254\u0281\u0292j\u0254\u0303/",
+  // place name (Dumas)
+  myriel: "/mi\u0281j\u025Bl/",
+  // character name (Hugo)
+  "neuve-sainte-genevi\xE8ve": "/n\u0153vs\u025B\u0303t\u0292\u0259nvj\u025Bv/",
+  // Paris street (Balzac)
+  nicole: "/nik\u0254l/",
+  pangloss: "/p\u0251\u0303\u0261l\u0254s/",
+  // Voltaire character
+  rainur\u00E9e: "/\u0281\u025Bny\u0281e/",
+  rion: "/\u0281j\u0254\u0303/",
+  // place name (Dumas)
+  "saint-marcel": "/s\u025B\u0303ma\u0281s\u025Bl/",
+  // Paris quarter (Dumas)
+  "thunder-ten-tronckh": "/t\u0254n\u025B\u0281t\u025Bnt\u0281\u0254nk/",
+  // Voltaire
+  trieste: "/t\u0281ij\u025Bst/",
+  // city name (Dumas)
+  vauquer: "/voke/",
+  // character name (Balzac)
+  y: "/i/"
+  // pronoun "there" — dict has letter name /igʁɛk/
+};
+
+// src/overrides/is.ts
+var is = {
+  \u00E1sir: "/\u02C8au\u02D0s\u026Ar/",
+  \u00E1synjur: "/\u02C8au\u02D0s\u026Anj\u028Fr/",
+  ballir: "/\u02C8patl\u026Ar/",
+  beytils: "/\u02C8pei\u02D0t\u02B0\u026Als/",
+  Bj\u00E1lfa: "/\u02C8pjaulva/",
+  Bjarnar: "/\u02C8pjartnar/",
+  bl\u00F3t: "/plou\u02D0t\u02B0/",
+  // sacrifice/ritual
+  b\u00F3klaus: "/\u02C8pouk\u02B0l\u0153ys/",
+  br\u00E9fberi: "/\u02C8prj\u025B\u02D0vp\u025Br\u026A/",
+  // letter carrier
+  burluf\u00F3ts: "/\u02C8p\u028Frtl\u028Ffouts/",
+  efla\u00F0i: "/\u02C8\u025Bpla\xF0\u026A/",
+  // past tense: strengthened/performed
+  ek: "/\u025Bk\u02B0/",
+  fannhv\u00EDtir: "/\u02C8fan\u02D0kvit\u02B0\u026Ar/",
+  farandi: "/\u02C8farant\u026A/",
+  fars\u00E6lda: "/\u02C8farsailta/",
+  fira: "/\u02C8f\u026Ara/",
+  flatnefur: "/\u02C8flatn\u025Bv\u028Fr/",
+  fl\u00FD\u00F0u: "/\u02C8fli\u02D0\xF0\u028F/",
+  fornaldar: "/\u02C8f\u0254rtnaltar/",
+  fornrit: "/\u02C8f\u0254rtnr\u026At\u02B0/",
+  galdursmenn: "/\u02C8kalt\u028Frsm\u025Bn\u02D0/",
+  gautr: "/\u02C8k\u0153y\u02D0tr/",
+  // New saga/literary overrides
+  gengr: "/k\u025B\u014Bkr/",
+  go\u00F0or\u00F0sma\u00F0ur: "/\u02C8k\u0254\u02D0\xF0\u0254r\xF0sma\u02D0\xF0\u028Fr/",
+  h\u00E6ngs: "/\u02C8hai\u014Bs/",
+  hags\u00E6lda: "/\u02C8haksailta/",
+  H\u00E1lfdanarsonar: "/\u02C8haulf\u02CCtanar\u02CCs\u0254nar/",
+  h\u00E1lftr\u00F6lls: "/\u02C8haulftr\u0153tls/",
+  Hallbjarnar: "/\u02C8hatlpjartnar/",
+  Hallfre\u00F0s: "/\u02C8hatlfr\u025B\xF0s/",
+  // genitive of Hallfreður
+  Hallfre\u00F0ur: "/\u02C8hatlfr\u025B\xF0\u028Fr/",
+  // proper name
+  h\u00E1rfagra: "/\u02C8haur\u02CCfa\u0263ra/",
+  // fair-haired (epithet)
+  h\u00E1rfagri: "/\u02C8haur\u02CCfa\u0263r\u026A/",
+  hersir: "/\u02C8h\u025Brs\u026Ar/",
+  hersis: "/\u02C8h\u025Brs\u026As/",
+  herskarar: "/\u02C8h\u025Brsk\u02B0arar/",
+  Hrafnkelsdal: "/\u02C8r\u0325apnk\u02B0\u025Blstalr/",
+  // place name
+  Hreggvi\u00F0sson: "/\u02C8r\u0325\u025Bk\u02D0v\u026A\xF0s\u02D0\u0254n/",
+  hr\u00EDmhv\u00EDta: "/\u02C8r\u0325imkvit\u02B0a/",
+  Ing\u00F3lfr: "/\u02C8\u026A\u014Bkoulfr/",
+  ins: "/\u026Ans/",
+  \u00CDvarssonar: "/\u02C8ivars\u02D0\u0254nar/",
+  j\u00F6klanna: "/\u02C8j\u0153k\u02B0lan\u02D0a/",
+  J\u00F3rsalalands: "/\u02C8jourtsalalants/",
+  Ketils: "/\u02C8k\u02B0\u025Bt\u02B0\u026Als/",
+  kindir: "/\u02C8k\u02B0\u026Ant\u026Ar/",
+  K\u00F3lumkilli: "/\u02C8k\u02B0ou\u02D0l\u028Fmk\u02B0\u026Atl\u026A/",
+  konungd\u00E6mi: "/\u02C8k\u02B0\u0254\u02D0n\u028F\u014B\u02CCtai\u02D0m\u026A/",
+  konungd\u00F3mur: "/\u02C8k\u02B0\u0254\u02D0n\u028F\u014B\u02CCdou\u02D0m\u028Fr/",
+  konungr: "/\u02C8k\u02B0\u0254\u02D0n\u028F\u014Bkr/",
+  konungseigninni: "/\u02C8k\u02B0\u0254\u02D0n\u028F\u014Bkseikn\u026An\u02D0\u026A/",
+  kotb\u00F3ndi: "/\u02C8k\u02B0\u0254t\u02B0pount\u026A/",
+  kunnigt: "/\u02C8k\u02B0\u028Fn\u02D0\u026Ak\u02B0t\u02B0/",
+  kvonga\u00F0ur: "/\u02C8k\u02B0v\u0254\u014Bka\xF0\u028Fr/",
+  kynst\u00F3r: "/\u02C8k\u02B0\u026Anstou\u02D0r/",
+  ma\u00F0r: "/ma\u02D0\xF0r/",
+  magra: "/\u02C8ma\u0263ra/",
+  mannd\u00E1\u00F0in: "/\u02C8man\u02D0tau\u02D0\xF0\u026An/",
+  mannf\u00F3lkit: "/\u02C8man\u02D0fouk\u02B0\u026At\u02B0/",
+  mj\u00F6k: "/mj\u0153k\u02B0/",
+  m\u00F6gu: "/\u02C8m\u0153\u02D0k\u028F/",
+  n\u00F3rr\u00E6nn: "/\u02C8nou\u02D0rrai\u02D0n\u02D0/",
+  N\u00F6rvasundum: "/\u02C8n\u0153rvas\u028Fnt\u028Fm/",
+  Norvegi: "/\u02C8n\u0254rv\u025Bj\u026A/",
+  n\u00FDu: "/\u02C8ni\u02D0\u028F/",
+  \u00F3arga: "/\u02C8ou\u02D0arka/",
+  \u00D6nundur: "/\u02C8\u0153\u02D0n\u028Fnt\u028Fr/",
+  or: "/\u0254r/",
+  \u00F3r: "/ou\u02D0r/",
+  \u00F6xn: "/\u02C8\u0153ksn/",
+  pl\u00F3gsland: "/\u02C8p\u02B0lou\u02D0kslant/",
+  r\u00E1\u00F0u: "/\u02C8rau\u02D0\xF0\u028F/",
+  r\u00E6kir: "/\u02C8rai\u02D0k\u02B0\u026Ar/",
+  Raumar\u00EDki: "/\u02C8r\u0153y\u02D0mar\u026A\u02D0k\u02B0\u026A/",
+  Raumsd\u00E6lafylki: "/\u02C8r\u0153ymstailaf\u026Alk\u02B0\u026A/",
+  Raumsdal: "/\u02C8r\u0153ymstal/",
+  raumur: "/\u02C8r\u0153y\u02D0m\u028Fr/",
+  Rein: "/rei\u02D0n/",
+  sannliga: "/\u02C8san\u02D0l\u026A\u0263a/",
+  Sk\u00ED\u00F0ason: "/\u02C8ski\u02D0\xF0as\u0254n/",
+  s\u00F6\u00F0ul: "/\u02C8s\u0153\u02D0\xF0\u028Fl/",
+  s\u00F3lkerfum: "/\u02C8soulk\u02B0\u025Brv\u028Fm/",
+  tindar: "/\u02C8t\u02B0\u026Antar/",
+  \u00FAtsj\u00E1num: "/\u02C8u\u02D0t\u02B0sjau\u02D0n\u028Fm/",
+  v\u00E1gskorin: "/\u02C8vau\u0263sk\u0254r\u026An/",
+  Valf\u00F6\u00F0r: "/\u02C8valv\u0153\xF0r/",
+  v\u00EDkverskur: "/\u02C8vi\u02D0kv\u025Brsk\u028Fr/",
+  Yngveldi: "/\u02C8\u026A\u014Bkv\u025Blt\u026A/",
+  \u00FEat: "/\u03B8at\u02B0/"
+};
+
+// src/overrides/ja.ts
+var ja = {
+  \u3042\u3051\u307C\u306E: "/akebono/",
+  \u3042\u3063\u305F: "/at\u02D0a/",
+  \u3042\u3068: "/ato/",
+  \u3042\u307E\u305F: "/amata/",
+  \u3042\u3089\u306C: "/a\u027Ean\u026F/",
+  \u3042\u308A: "/a\u027Ei/",
+  \u3042\u308B: "/a\u027E\u026F/",
+  \u3042\u308B\u304F: "/a\u027E\u026Fk\u026F/",
+  \u3044\u3046: "/i\u026F/",
+  \u30A4\u30AE\u30EA\u30B9: "/i\u0261i\u027Eis\u026F/",
+  // England
+  \u3044\u305F: "/ita/",
+  \u3044\u305F\u3046: "/ita\u026F/",
+  // archaic: greatly
+  // New sample overrides
+  \u3044\u3064\u3082: "/its\u026Fmo/",
+  \u3044\u3065\u308C: "/id\u026F\u027Ee/",
+  // archaic: which
+  \u3044\u308B: "/i\u027E\u026F/",
+  \u3046\u307F: "/\u026Fmi/",
+  \u3046\u3089: "/\u026F\u027Ea/",
+  \u304A\u308B: "/o\u027E\u026F/",
+  \u304B\u305F: "/kata/",
+  \u304B\u305F\u3061: "/katat\u0255i/",
+  \u304B\u3064\u3050: "/kats\u026F\u0261\u026F/",
+  \u304B\u306E: "/kano/",
+  \u304F: "/k\u026F/",
+  // archaic verb stem
+  \u3054: "/\u0261o/",
+  \u3053\u3053: "/koko/",
+  \u3054\u3056\u3044\u307E\u3059: "/\u0261ozaimas\u026F/",
+  \u3053\u306E: "/kono/",
+  \u3055\u3056\u308C: "/saza\u027Ee/",
+  \u3055\u3073\u3057\u3044: "/sabi\u0255i\u02D0/",
+  \u3055\u3089: "/sa\u027Ea/",
+  // archaic: furthermore
+  \u3057\u3088\u3046: "/\u0255ijo\u02D0/",
+  \u3059\u3050\u308C\u308B: "/s\u026F\u0261\u026F\u027Ee\u027E\u026F/",
+  \u3059\u3053\u3057: "/s\u026Fko\u0255i/",
+  \u3059\u308B: "/s\u026F\u027E\u026F/",
+  \u305D\u3053: "/soko/",
+  \u305D\u306E: "/sono/",
+  \u3060\u3044\u3076: "/daib\u026F/",
+  \u3060\u3051: "/dake/",
+  \u305F\u3060: "/tada/",
+  \u3060\u3064: "/dats\u026F/",
+  // archaic suffix: -ish
+  \u305F\u306A\u3073\u304F: "/tanabik\u026F/",
+  \u3064\u304F: "/ts\u026Fk\u026F/",
+  \u3064\u308C\u308B: "/ts\u026F\u027Ee\u027E\u026F/",
+  \u3069\u3053: "/doko/",
+  \u3068\u3053\u308D: "/toko\u027Eo/",
+  \u3068\u3089\u3048\u308B: "/to\u027Eae\u027E\u026F/",
+  \u306A\u304A: "/nao/",
+  \u306A\u304B: "/naka/",
+  \u306A\u304F: "/nak\u026F/",
+  \u306A\u3063\u305F: "/nat\u02D0a/",
+  \u306A\u3073\u304F: "/nabik\u026F/",
+  \u306E\u3051: "/noke/",
+  \u3070\u304B\u308A: "/baka\u027Ei/",
+  \u306F\u305F: "/hata/",
+  \u30CF\u30F3\u30D6\u30EB\u30AF: "/hamb\u026F\u027E\u026Fk\u026F/",
+  \u3075\u3046: "/\u0278\u026F\u02D0/",
+  \u3075\u3061: "/\u0278\u026Ft\u0255i/",
+  \u307B\u304B: "/hoka/",
+  \u307B\u3069: "/hodo/",
+  \u307B\u3093\u3068\u3046: "/honto\u02D0/",
+  \u307E\u305F: "/mata/",
+  \u307E\u3060: "/mada/",
+  \u307E\u3067: "/made/",
+  \u307F\u306A\u3055\u3093: "/minasa\u0274/",
+  \u307F\u3093\u306A: "/min\u02D0a/",
+  \u3080\u3059: "/m\u026Fs\u026F/",
+  \u30E1\u30ED\u30B9: "/me\u027Eos\u026F/",
+  \u3084\u3046\u3084\u3046: "/ja\u026Fja\u026F/",
+  // archaic: gradually
+  \u3084\u307F: "/jami/",
+  \u3084\u3080: "/jam\u026F/",
+  \u3084\u3093\u3054\u3068\u306A\u3057: "/ja\u014B\u0261otona\u0255i/",
+  // archaic: noble
+  \u3088\u304F: "/jok\u026F/",
+  \u3088\u307B\u3069: "/johodo/",
+  \u308B: "/\u027E\u026F/",
+  // classical auxiliary
+  \u308F\u304B\u308B: "/waka\u027E\u026F/",
+  \u4E00\u8336: "/it\u02D0\u0255a/",
+  // Issa (poet)
+  \u4E09\u5341\u4E03: "/sa\u0274d\u0291\u026F\u02D0\u0255it\u0255i/",
+  \u5019\u3046: "/so\u02D0\u027Eo\u02D0/",
+  // archaic polite auxiliary
+  \u547C\u3093\u3067: "/jo\u0274de/",
+  \u6B62\u307E\u3063\u305F: "/tomat\u02D0a/",
+  \u75E9: "/jase/",
+  // thin
+  \u767D\u304F: "/\u0255i\u027Eok\u026F/",
+  \u77E5\u3063\u3066: "/\u0255it\u02D0e/",
+  \u7A4D\u307F: "/ts\u026Fmi/",
+  \u7F85\u751F\u9580: "/\u027Ea\u0255o\u02D0mo\u0274/",
+  // Rashomon
+  \u821E\u9DB4: "/maiz\u026F\u027E\u026F/",
+  // Maizuru (place)
+  \u91D1\u95A3: "/ki\u0274kak\u026F/",
+  // Golden Pavilion
+  \u9759\u304B\u3055: "/\u0255iz\u026Fkasa/"
+};
+
+// src/overrides/km.ts
+var km = {
+  // --- UDHR Article 1 ---
+  \u1780\u17C6\u178E\u17BE\u178F: "/k\u0251mna\u0259t/",
+  // birth
+  // --- Reamker (Ramayana) ---
+  \u1780\u17D2\u178A\u17B8: "/kd\u0259y/",
+  // matter, affair
+  \u1780\u17D2\u179A\u17C1\u179C: "/kreew/",
+  // furious
+  \u1780\u17D2\u179A\u17C1\u179C\u1780\u17D2\u179A\u17C4\u1792: "/kreew krout/",
+  // furiously angry
+  \u1780\u17D2\u179A\u17C4\u1792: "/krout/",
+  // anger
+  \u1780\u17D2\u179F: "/ks/",
+  // consonant cluster (browser splits ក្សត្រី)
+  \u1780\u17D2\u179F\u178F\u17D2\u179A: "/ksaat/",
+  // king
+  // --- Nokor Reach (National Anthem) ---
+  \u1780\u17D2\u179F\u178F\u17D2\u179A\u17B6: "/ksaatraa/",
+  // king (literary form)
+  \u1780\u17D2\u179F\u178F\u17D2\u179A\u17B8: "/ksaatr\u0259y/",
+  // queen
+  \u1781\u17D2\u1798\u17B8: "/km\u0259y/",
+  // ogre (literary)
+  \u1781\u17D2\u179C\u17BE\u1780: "/kwa\u0259k/",
+  // to stir, disturb
+  \u1782\u17B6\u1794\u17CB: "/koap/",
+  // beloved
+  // --- Tum Teav ---
+  \u1782\u17BC: "/kuu/",
+  // partner
+  \u1782\u17BC\u1782\u17B6\u1794\u17CB: "/kuu koap/",
+  // partner, beloved
+  \u1783\u17D2\u179B\u17B6\u178F: "/kliet/",
+  // separated
+  \u1785\u179A: "/c\u0251\u0251/",
+  // to walk, go
+  \u1785\u17D2\u1793\u17C1\u17C7: "/cneh/",
+  // this (literary)
+  \u1787\u17D0\u1799: "/cey/",
+  // victory
+  \u1787\u17D0\u1799\u1798\u1784\u17D2\u1782\u179B: "/cey m\u0254\u014Bk\u0254l/",
+  // victory, auspicious
+  // --- Proverbs ---
+  \u1787\u17B6\u1780\u17CB: "/ceak/",
+  // certain, sure
+  \u1787\u17B6\u179B\u17C6\u178A\u17B6\u1794\u17CB: "/cie l\u0254m\u0257aap/",
+  // successively
+  // --- Preah Chinawong ---
+  \u1787\u17B7\u1793: "/c\u0268n/",
+  // Jin (proper name element)
+  \u1787\u17B7\u1793\u179C\u1784\u17D2\u0E2A: "/c\u0268n w\u0254\u014B/",
+  // Chinawong (proper name)
+  \u1787\u17BC: "/cuu/",
+  // proper (literary)
+  \u1789\u17B6\u1780\u17CB: "/\u0272eak/",
+  // to startle, twitch
+  \u1789\u17B6\u1780\u17CB\u1785\u17B7\u1789\u17D2\u1785\u17BE\u1798: "/\u0272eak c\u0259\u0272ca\u0259m/",
+  // to raise eyebrows
+  \u178A\u17C2\u1793: "/daen/",
+  // territory
+  \u178E\u17B6\u1799: "/naay/",
+  // to yearn
+  \u178E\u17B6\u1799\u1785\u17B7\u178F\u17D2\u178F: "/naay c\u0259t/",
+  // heart yearns
+  \u178F\u1794: "/t\u0251p/",
+  // to reply
+  \u1790\u17D2\u1780\u17B6\u1793: "/tkaan/",
+  // magnificent (literary)
+  \u1790\u17D2\u1780\u17BE\u1784: "/tk\u0259\u0259\u014B/",
+  // to glorify
+  \u1790\u17D2\u1780\u17BE\u1784\u1790\u17D2\u1780\u17B6\u1793: "/tk\u0259\u0259\u014B tkaan/",
+  // glorious, magnificent
+  \u1790\u17D2\u1793\u17BC\u179A: "/tnoo/",
+  // dignity, nobility
+  \u1790\u17D2\u1793\u17C2: "/tnae/",
+  // aspect (partial segmentation of ថ្នែក)
+  \u1790\u17D2\u1793\u17C2\u1780: "/tnaek/",
+  // aspect, class
+  \u1790\u17D2\u179C\u17B6\u178F\u17CB: "/twat/",
+  // harshly
+  \u1791\u178F: "/t\u0254t/",
+  // to look, behold (royal)
+  \u1791\u17B6\u179C: "/tiew/",
+  // Teav (proper name)
+  \u1791\u17BB\u1780\u17D2\u1781: "/tuk/",
+  // suffering (dukkha)
+  \u1791\u17BB\u1780\u17D2\u1781\u179C\u17C1\u1791\u1793\u17B6: "/tukweet\u0251\u0251nie/",
+  // suffering
+  \u1791\u17BC\u179B: "/tuul/",
+  // to inform (royal register)
+  \u1791\u17C1\u1796\u17D2\u178F\u17B6: "/teep\u0257aa/",
+  // devas, celestial beings
+  \u1793\u179A\u1794\u178F\u17B8: "/n\u0254r\u0254pa\u0257\u0259y/",
+  // sovereign, king
+  \u1793\u17B6\u179C: "/niew/",
+  // Nav (proper name)
+  \u1793\u17B7\u1798\u17CC\u179B: "/n\u0268mm\u0254l/",
+  // pure, immaculate
+  \u1794\u17C6\u1795\u17D2\u179B\u17B6\u1789: "/b\u0251mp\u02B0lie\u0272/",
+  // to destroy
+  \u1794\u17C6\u1795\u17D2\u179B\u17B7\u1785: "/b\u0251mp\u02B0l\u0259c/",
+  // to demolish
+  \u1794\u1796\u17B7\u178F\u17D2\u179A: "/b\u0251p\u0268t/",
+  // lord, sir (polite address)
+  \u1794\u17BB\u178F\u17D2\u179A: "/\u0253ot/",
+  // son
+  // --- Chbab Srey (Code for Women) ---
+  \u1794\u17D2\u179A\u178A\u17C5: "/pr\u0251\u0257aw/",
+  // to advise, counsel
+  \u1794\u17D2\u179A\u178E\u17B8: "/pr\u0251n\u0259y/",
+  // loving, affectionate
+  \u1794\u17D2\u179A\u17B6\u1787\u17D2\u1789: "/praac/",
+  // wisdom
+  \u1794\u17D2\u179A\u17B6\u179F\u17B6\u1791: "/praasaat/",
+  // temple, palace
+  \u1796\u17B7\u179A\u17C4\u1792: "/piroot/",
+  // anger
+  \u1796\u17BB\u17C6: "/pum/",
+  // not (literary)
+  \u1796\u17BB\u17C6\u1787\u17BC: "/pum cuu/",
+  // not proper
+  \u1796\u17D2\u1799\u17B6\u1794\u17B6\u1791: "/pjiebaat/",
+  // malice
+  \u1796\u17D2\u179A\u17C7\u1798\u17A0\u17C1\u179F\u17B8: "/preah m\u0254hees\u0259y/",
+  // queen consort (with ព្រះ)
+  \u1796\u17D2\u179A\u17C7\u179A\u17B6\u1787\u1794\u17BB\u178F\u17D2\u179A: "/preah riec \u0253ot/",
+  // prince (with ព្រះ)
+  \u1796\u17D2\u179A\u17C7\u179A\u17B6\u1787\u17B6: "/preah riecie/",
+  // the king
+  \u1797\u17D0\u1780\u17D2\u179A\u17D2\u178F: "/p\u02B0eak/",
+  // face (royal register)
+  \u1797\u17B6\u178F\u179A: "/p\u02B0iet\u0251\u0251/",
+  // brother (standalone; final រ silent)
+  \u1797\u17B6\u178F\u179A\u1797\u17B6\u1796: "/p\u02B0iet\u0251\u0251r\u0251p\u02B0iep/",
+  // brotherhood
+  \u1797\u17D2\u1789\u17B6: "/p\u0272ie/",
+  // to awaken
+  \u1798\u1784\u17D2\u1782\u179B: "/m\u0254\u014Bk\u0254l/",
+  // auspicious
+  \u1798\u17A0\u17B6: "/m\u0254haa/",
+  // great (prefix)
+  \u1798\u17A0\u17B6\u1780\u17D2\u179F\u178F\u17D2\u179A: "/m\u0254haa ksaat/",
+  // great king, maharaja
+  \u1798\u17A0\u17B6\u1787\u17B6\u178F\u17B7: "/m\u0254haa ciet/",
+  // great nation
+  \u1798\u17A0\u17C1\u179F\u17B8: "/m\u0254hees\u0259y/",
+  // queen consort
+  \u179A\u17B6\u1787: "/riec/",
+  // royal
+  \u179A\u17B6\u1787\u1794\u17BB\u178F\u17D2\u179A: "/riec \u0253ot/",
+  // prince
+  \u179A\u17B6\u1787\u17B6: "/riecie/",
+  // king
+  \u179A\u17BB\u1784: "/ru\u014B/",
+  // to shine
+  \u179A\u17BB\u1784\u179A\u17BF\u1784: "/ru\u014B r\u0268\u0259\u014B/",
+  // brilliant, prosperous
+  \u179B\u17C6\u178A\u17B6\u1794\u17CB: "/l\u0254m\u0257aap/",
+  // successively, in order
+  \u179C\u1784\u17D2\u179F: "/w\u0254\u014B/",
+  // dynasty, lineage
+  \u179C\u17B7\u1785\u17B6\u179A\u178E\u1789\u17D2\u1789\u17B6\u178E: "/wicaar\u0251\u0272\u0272aan/",
+  // discernment
+  \u179C\u17C1\u1791\u1793\u17B6: "/weet\u0251\u0251nie/",
+  // pain (vedana)
+  \u179F\u178F\u17B7: "/sa\u0294te\u0294/",
+  // consciousness, mindfulness
+  \u179F\u178F\u17B7\u179F\u1798\u17D2\u1794\u1787\u1789\u17D2\u1789\u17C8: "/sa\u0294te\u0294 sampa\u0294c\u028A\u0259\u0272\u0272ea\u0294/",
+  // conscience
+  \u179F\u1798\u17D2\u1794\u1787\u1789\u17D2\u1789\u17C8: "/sampa\u0294c\u028A\u0259\u0272\u0272ea\u0294/",
+  // awareness
+  \u179F\u17B6\u1791\u179A: "/saat\u0254\u0254/",
+  // to welcome
+  \u179F\u17B7\u179A\u17B8: "/ser\u0259y/",
+  // glory (from Pali)
+  \u179F\u17BD: "/su\u0259/",
+  // (first syllable of សួស្តី)
+  \u179F\u17BD\u179F\u17D2\u178F\u17B8: "/su\u0259sd\u0259y/",
+  // greeting, well-being
+  \u179F\u17C1\u1785\u1780\u17D2\u178A\u17B8: "/sac kd\u0259y/",
+  // matter, affair
+  \u179F\u17C1\u1785\u1780\u17D2\u178A\u17B8\u1790\u17D2\u179B\u17C3\u1790\u17D2\u1793\u17BC\u179A: "/sac kd\u0259y tlay tnoo/",
+  // dignity
+  \u179F\u17D2\u178A\u17C1\u1785: "/sdac/",
+  // king
+  \u179F\u17D2\u178F\u17B6\u1794\u17CB: "/sdaap/",
+  // to listen
+  \u179F\u17D2\u178F\u17B8: "/sd\u0259y/",
+  // (second syllable of សួស្តី)
+  \u179F\u17D2\u1791\u17BB\u17C7: "/stuh/",
+  // to rush
+  \u179F\u17D2\u1793\u17C6: "/sn\u0251m/",
+  // concubine
+  \u17A0\u178F\u17D2\u1790\u17B6: "/hat\u0251\u0251t\u02B0aa/",
+  // hand (literary)
+  // --- Constitution Preamble ---
+  \u17A2\u1784\u17D2\u1782\u179A: "/\u0294\u0251\u014Bk\u0254\u0254/",
+  // Angkor
+  \u17A2\u179A: "/\u0294\u0251\u0251/",
+  // to rejoice
+  \u17A2\u179F\u17D2\u1785\u17B6\u179A\u17D2\u1799: "/\u0294\u0251scaa/",
+  // wonderful, marvelous
+  \u17A2\u17B6\u179B: "/\u0294aal/",
+  // to rush, boast
+  \u17A2\u17BD\u178F: "/\u0294u\u0259t/",
+  // to boast
+  \u17B1\u17D2\u1799: "/\u0294aoy/"
+  // to give, let
+};
+
+// src/overrides/ko.ts
+var ko = {
+  \uAC10\uB098\uBB34: "/kam.na.mu/",
+  \uAC1C\uCC9C: "/k\u025B.t\u0255\u02B0\u028Cn/",
+  \uACBD\uC131: "/kj\u028C\u014B.s\u028C\u014B/",
+  // old name for Seoul
+  // New sample overrides
+  \uACE0\uB2EC\uD504\uB2E4: "/ko.dal.p\u02B0\u026F.da/",
+  \uACE0\uC774: "/ko.i/",
+  \uAD11\uC74C: "/kwa\u014B.\u026Fm/",
+  \uAE08\uBE5B: "/k\u026Fm.pit\u031A/",
+  \uAE30\uB098\uAE30\uB2E4: "/ki.na.\u0261i.da/",
+  // very long
+  \uAE38\uB3D9: "/kil.do\u014B/",
+  // Hong Gildong (character name)
+  \uAF79\uACFC\uB9AC: "/k\u0348w\u025B\u014B.gwa.\u027Ei/",
+  // kkwaenggwari (percussion instrument)
+  \uB04A\uC784: "/k\u0348\u026Fn.im/",
+  \uB098\uD0C0\uC0E4: "/na.t\u02B0a.\u0255a/",
+  // Natasha
+  \uB17C\uAC00: "/non.\u0261a/",
+  \uB2C8\uAE4C: "/ni.k\u0348a/",
+  // because (suffix)
+  \uB3C4\uC6B0\uB2E4: "/to.u.da/",
+  // to help
+  \uB3D9\uC9D3\uB2EC: "/to\u014B.d\u0291it\u031A.t\u0348al/",
+  \uB9E4\uC5B4\uB2EC\uB9AC\uB2E4: "/m\u025B.\u028C.dal.li.da/",
+  \uBB3C\uB4E4\uB2E4: "/mul.d\u026Fl.da/",
+  \uBC31\uB450\uC0B0: "/p\u025Bk\u031A.t\u0348u.san/",
+  // Mt. Baekdu
+  \uBC94\uD558\uB2E4: "/p\u028Cm.ha.da/",
+  \uC0B0\uBAA8\uD241\uC774: "/san.mo.t\u02B0u\u014B.i/",
+  \uC18C\uC90F\uC9D1: "/so.d\u0291ut\u031A.t\u0348\u0255ip\u031A/",
+  \uC544\uB77C\uB9AC\uC694: "/a.\u027Ea.\u027Ei.jo/",
+  // arirang refrain
+  \uC544\uC2DC\uB2E4: "/a.\u0255i.da/",
+  // to know (honorific)
+  \uC57D\uC0B0: "/jak\u031A.s\u0348an/",
+  // Yaksan (place)
+  \uC5B4\uB450: "/\u028C.du/",
+  \uC5B4\uB860: "/\u028C.\u027Eon/",
+  // archaic: elder
+  \uC5ED\uACB9\uB2E4: "/j\u028Ck\u031A.kj\u028Cp\u031A.t\u0348a/",
+  \uC601\uBCC0: "/j\u028C\u014B.bj\u028Cn/",
+  // Yeongbyeon (place)
+  \uC624\uB3D9\uB098\uBB34: "/o.do\u014B.na.mu/",
+  \uC624\uB3D9\uC78E: "/o.do\u014B.ip\u031A/",
+  \uC624\uC2DC\uB2E4: "/o.\u0255i.da/",
+  // honorific: to come
+  \uC654\uB2E4: "/wat\u031A.t\u0348a/",
+  \uC6B0\uB7EC\uB974\uB2E4: "/u.\u027E\u028C.\u027E\u026F.da/",
+  \uC6D0\uD1B5\uD558\uB2E4: "/w\u028Cn.t\u02B0o\u014B.ha.da/",
+  \uC774\uB77C: "/i.\u027Ea/",
+  // copula ending
+  \uC774\uC5B4\uB4E0: "/i.\u028C.d\u026Fn/",
+  // archaic conditional
+  \uC774\uC5D0: "/i.e/",
+  \uC78A\uD788\uB2E4: "/i.t\u02B0i.da/",
+  \uC78E\uC0C8: "/ip\u031A.s\u0348\u025B/",
+  \uC7A5\uB0A0: "/t\u0255a\u014B.nal/",
+  \uC7A5\uD130: "/t\u0255a\u014B.t\u02B0\u028C/",
+  \uC810\uC21C\uC774: "/t\u0255\u028Cm.su.ni/",
+  // character name (diminutive)
+  \uC9C0\uB9AC\uB2E4: "/t\u0255i.\u027Ei.da/",
+  \uC9C0\uC904\uB300\uB2E4: "/t\u0255i.d\u0291ul.d\u025B.da/",
+  \uCC44\uC2DD: "/t\u0255\u02B0\u025B.\u0255ik\u031A/",
+  \uCCAD\uCC9C: "/t\u0255\u02B0\u028C\u014B.t\u0255\u02B0\u028Cn/",
+  \uCD5C: "/t\u0255\u02B0we/",
+  // surname Choi
+  \uCD98\uD48D: "/t\u0255\u02B0un.p\u02B0u\u014B/",
+  \uD0C0\uC791\uB9C8\uB2F9: "/t\u02B0a.d\u0291ak\u031A.ma.da\u014B/",
+  // threshing ground
+  \uD2F0\uB04C: "/t\u02B0i.k\u0348\u026Fl/",
+  \uD53C\uC5B4\uC624\uB974\uB2E4: "/p\u02B0i.\u028C.o.\u027E\u026F.da/",
+  \uD55C\uD14C: "/han.t\u02B0e/",
+  \uD574\uC124\uD53C: "/h\u025B.s\u028Cl.p\u02B0i/",
+  // poetic: at sunset
+  \uD5E4\uB2E4: "/he.da/",
+  // archaic: to count
+  \uD718\uB2EC\uB9AC\uB2E4: "/hwi.dal.li.da/",
+  \uD718\uB3CC\uB2E4: "/hwi.dol.da/"
+};
+
+// src/overrides/ma.ts
+var ma = {
+  Abdul: "abdul",
+  bebas: "bebas",
+  boleh: "boleh",
+  gemawan: "\u0261\u0259mawan",
+  Hamid: "hamid",
+  ibni: "ibni",
+  Jebat: "d\u0292\u0259bat",
+  Johor: "d\u0292oho\u027E",
+  kebebasan: "k\u0259bebasan",
+  Lekir: "l\u0259ki\u027E",
+  Lekiu: "l\u0259kiw",
+  Malaysia: "malejsia",
+  merdeka: "m\u0259\u027Edeka",
+  mereka: "m\u0259reka",
+  Oleh: "oleh",
+  samarata: "samarata",
+  Sarawak: "sa\u027Eawak",
+  seekor: "s\u0259eko\u027E",
+  seksaan: "seksa\u0294an",
+  Selangor: "s\u0259la\u014Bo\u027E",
+  selendang: "s\u0259l\u0259nda\u014B",
+  Singapura: "si\u014B\u0261apu\u027Ea",
+  Terengganu: "t\u0259\u027Ee\u014B\u0261anu",
+  Tuhan: "tuhan"
+};
+
+// src/overrides/nb.ts
+var nb = {
+  // Old orthography (Riksmål/Danish) and common words
+  al: "\u0251\u02D0l",
+  \u00E5rsalderen: "o\u02D0\u027Es\u0251l\u02D0d\u0259\u027E\u0259n",
+  arv: "\u0251\u027Ev",
+  behold: "b\u0259h\u0254l",
+  Bj\u00F8rgulfson: "bj\xF8\u02D0\u027E\u0261\u0289lfs\u0254n",
+  Blik: "bl\u026Ak",
+  Bliv: "bli\u02D0",
+  b\u00F8r: "b\xF8\u02D0\u027E",
+  B\u00F8rnene: "b\xF8\u02D0\u0273\u0259n\u0259",
+  brorskapets: "b\u027Eu\u02D0\u027Esk\u0251\u02D0p\u0259ts",
+  demokratiet: "d\u025Bmok\u027E\u0251ti\u02D0\u0259",
+  dig: "d\u0251j",
+  Dyb: "dy\u02D0p",
+  Eftermiddag: "\u025Bft\u025B\u027Em\u026Ad\u0251\u02D0\u0261",
+  ej: "\u0251j",
+  fandens: "f\u0251n\u02D0\u0259ns",
+  f\u00E5tt: "f\u0254t",
+  fiender: "fi\u02D0\u025Bnd\u0259\u027E",
+  Fjeldbygden: "fj\u025Blb\u028F\u0261d\u0259n",
+  fordrukken: "f\u0254\u027Ed\u027E\u0289k\u02D0\u0259n",
+  fornuft: "f\u0254\u027En\u0289ft",
+  frem: "f\u027E\u025Bm",
+  Fremmedkarl: "f\u027E\u025Bm\u02D0\u0259k\u0251\u027El",
+  frostblaa: "f\u027E\u0254stblo\u02D0",
+  frygt: "f\u027E\u028Fkt",
+  gamlingen: "\u0261\u0251ml\u026A\u014B\u0259n",
+  Gem: "j\u025Bm",
+  gikk: "j\u026Ak",
+  Gjesling: "j\u025Bsl\u026A\u014B",
+  gjorde: "ju\u02D0\u027E\u0259",
+  Glands: "\u0261l\u0251ns",
+  gnistred: "\u0261n\u026Ast\u027E\u0259d",
+  gr\u00E5spr\u00E6ngt: "\u0261\u027Eo\u02D0sp\u027E\u025B\u014Bt",
+  Grunnlov: "\u0261\u027E\u0289n\u02D0lo\u02D0v",
+  Guld: "\u0261\u0289l",
+  Gyldenlak: "j\u028Fl\u02D0\u0259nl\u0251k",
+  Gyldentop: "j\u028Fl\u02D0\u0259nt\u0254p",
+  ham: "h\u0251m",
+  Hej: "h\u0251j",
+  h\u00F8r: "h\xF8\u02D0\u027E",
+  Horisonten: "h\u0254\u027E\u026As\u0254nt\u0259n",
+  humanistiske: "h\u0289m\u0251n\u026Ast\u026Ask\u0259",
+  husbond: "h\u0289\u02D0sb\u0254n",
+  Hvad: "v\u0251",
+  hvem: "v\u025Bm",
+  hverandre: "v\u025B\u027E\u0251nd\u027E\u0259",
+  hverken: "v\u025B\u027Ek\u0259n",
+  Hvor: "vu\u02D0\u027E",
+  hvoraf: "vu\u02D0\u027E\u0251v",
+  Hvorfor: "v\u0254\u027Ef\u0254\u027E",
+  Hvormeget: "vu\u02D0\u027Em\u0251j\u02D0\u0259",
+  iaften: "i\u0251ft\u0259n",
+  Idet: "i\u02D0de\u02D0t",
+  ihob: "iho\u02D0b",
+  imellem: "im\u025Bl\u02D0\u0259m",
+  imod: "i\u02D0mu\u02D0t",
+  inn: "\u026An",
+  intet: "\u026Ant\u0259",
+  Intet: "\u026Ant\u0259",
+  Ivar: "i\u02D0v\u0251\u027E",
+  Jammer: "j\u0251m\u025B\u027E",
+  Jensen: "j\u025Bns\u0259n",
+  jordegods: "ju\u02D0\u027E\u0259\u0261\u0254ts",
+  juletr\u00E6et: "j\u0289\u02D0l\u0259t\u027E\u025B\u02D0\u0259",
+  just: "j\u0289st",
+  kan: "k\u0251n",
+  kanske: "k\u0251n\u0283\u0259",
+  Kj\u00F8ttmeisene: "\xE7\xF8tm\xE6\u026As\u0259n\u0259",
+  kold: "k\u0254l",
+  kom: "k\u0254m",
+  Kongeriket: "k\u0254\u014B\u0259\u027Ei\u02D0k\u0259",
+  Kringsatt: "k\u027E\u026A\u014Bs\u0251t",
+  Kristiania: "k\u027E\u026Ast\u026A\u0251\u02D0n\u026A\u0251",
+  kristne: "k\u027E\u026Astn\u0259",
+  Landets: "l\u0251n\u0259ts",
+  Lavrans: "l\u0251\u02D0v\u027E\u0251ns",
+  lektor: "l\u025Bkt\u0254\u027E",
+  lig: "li\u02D0\u0261",
+  m\u00E5: "mo\u02D0",
+  maa: "mo\u02D0",
+  Mandemagt: "m\u0251n\u02D0\u0259m\u0251kt",
+  menneskerettighetene: "m\u025Bn\u02D0\u0259sk\u0259\u027E\u025Bt\u02D0\u026A\u0261he\u02D0t\u0259n\u0259",
+  menneskerettigheter: "m\u025Bn\u02D0\u0259sk\u0259\u027E\u025Bt\u02D0\u026A\u0261he\u02D0t\u0259\u027E",
+  menneskeverd: "m\u025Bn\u02D0\u0259sk\u0259ve\u02D0\u027Ed",
+  mig: "m\u0251j",
+  mod: "mo\u02D0d",
+  monarkisk: "m\u0254n\u0251\u027Ek\u026Ask",
+  Muld: "m\u0289l",
+  Nej: "n\u0251j",
+  noget: "no\u02D0\u0259",
+  n\u00F8gne: "n\xF8jn\u0259",
+  Norge: "n\u0254\u027E\u0261\u0259",
+  nu: "n\u0289\u02D0",
+  nysn\u00F8en: "ny\u02D0sn\xF8\u02D0\u0259n",
+  \u00F8jne: "\u0254jn\u0259",
+  op: "\u0254p",
+  Peer: "pe\u02D0\u027E",
+  Plads: "pl\u0251s",
+  proppenerer: "p\u027E\u0254p\u0259ne\u02D0\u027E\u0259\u027E",
+  Puslinger: "p\u0289\u02D0sl\u026A\u014B\u0259\u027E",
+  raaber: "\u027Eo\u02D0b\u0259\u027E",
+  R\u00E6kke: "\u027E\u025Bk\u0259",
+  r\u00E6kker: "\u027E\u025Bk\u02D0\u0259\u027E",
+  Ragnfrid: "\u027E\u0251\u014Bnf\u027Ei\u02D0d",
+  regjeringsform: "\u027Eeje\u02D0\u027E\u026A\u014Bsf\u0254\u027Em",
+  rettsstaten: "\u027E\u025Bt\u02D0s\u02D0t\u0251\u02D0t\u0259n",
+  Rosentinter: "\u027Eu\u02D0s\u0259nt\u026Ant\u025B\u027E",
+  rummeligt: "\u027E\u0289m\u02D0\u0259l\u026A\u0261t",
+  Ryg: "\u027E\u028F\u0261",
+  saganatt: "s\u0251\u02D0\u0261\u0251n\u0251t",
+  samvittighet: "s\u0251mv\u026At\u02D0\u026A\u0261he\u02D0t",
+  sidste: "s\u026Ast\u0259",
+  siger: "si\u02D0\u0259\u027E",
+  skabt: "sk\u0251pt",
+  skal: "sk\u0251l",
+  Skar: "sk\u0251\u02D0\u027E",
+  skj\u00E6ms: "\u0283\u025Bms",
+  Skodden: "sk\u0254d\u02D0\u0259n",
+  Skydsskiftet: "\u0283\u028Ftssk\u026Aft\u0259",
+  slig: "sli\u02D0\u0261",
+  smaa: "smo\u02D0",
+  Snees: "sne\u02D0s",
+  Sneskavler: "sne\u02D0sk\u0251\u02D0vl\u025B\u027E",
+  S\u00F8rby: "s\xF8\u02D0\u027Eby\u02D0",
+  sp\u00F8r: "sp\xF8\u02D0\u027E",
+  steget: "ste\u02D0\u0261\u0259",
+  stundom: "st\u0289nd\u0254m",
+  stygt: "st\u028Fkt",
+  Sundbu: "s\u0289nb\u0289\u02D0",
+  Syd: "sy\u02D0d",
+  tabt: "t\u0251pt",
+  Tant: "t\u0251nt",
+  Terje: "t\u025B\u027Ej\u0259",
+  T\u00F8v: "t\xF8\u02D0v",
+  Tvi: "tvi\u02D0",
+  tykkner: "t\u028Fk\u02D0n\u0259\u027E",
+  uavhendelig: "\u0289\u02D0\u0251vh\u025Bnd\u0259l\u026A\u0261",
+  udekket: "\u0289\u02D0d\u025Bk\u02D0\u0259",
+  udelelig: "\u0289\u02D0de\u02D0l\u0259\u0261",
+  uden: "\u0289\u02D0d\u0259n",
+  underjordisk: "\u0289n\u02D0\u0259\u027Eju\u02D0\u027Ed\u026Ask",
+  v\u00E6rbitt: "v\xE6\u02D0\u027Eb\u026At",
+  vejr: "ve\u02D0\u027E",
+  Verdigrunnlaget: "ve\u02D0\u027Ed\u026A\u0261\u027E\u0289n\u02D0l\u0251\u02D0\u0261\u0259",
+  Vigen: "vi\u02D0\u0261\u0259n",
+  vil: "v\u026Al",
+  Vindvet: "v\u026Andv\u0259",
+  visst: "v\u026Ast",
+  Vorherres: "vo\u02D0\u027Eh\u025B\u027E\u0259s",
+  yderste: "y\u02D0d\u0259\u027Est\u0259"
+};
+
+// src/overrides/nl.ts
+var nl = {
+  aandelen: "/\u02C8a\u02D0nd\u0259l\u0259n/",
+  achterhaalt: "/\u02C8\u0251xt\u0259rha\u02D0lt/",
+  ademde: "/\u02C8a\u02D0d\u0259md\u0259/",
+  // past tense: breathed
+  alchemie: "/\u0251lx\u0259\u02C8mi/",
+  allen: "/\u02C8\u0251l\u0259n/",
+  alsoo: "/\u0251l\u02C8so\u02D0/",
+  // archaic: thus
+  Anton: "/\u02C8\u0251nt\u0254n/",
+  // proper name
+  balkons: "/b\u0251l\u02C8k\u0254ns/",
+  bange: "/\u02C8b\u0251\u014B\u0259/",
+  benaeuwde: "/b\u0259\u02C8na\u02D0ud\u0259/",
+  // archaic: oppressed
+  // New sample overrides
+  benoemd: "/b\u0259\u02C8nu\u02D0mt/",
+  bevonden: "/b\u0259\u02C8v\u0254nd\u0259n/",
+  bleeke: "/\u02C8ble\u02D0k\u0259/",
+  // archaic: pale
+  bloedroze: "/\u02C8blud\u02CCro\u02D0z\u0259/",
+  boomen: "/\u02C8bo\u02D0m\u0259n/",
+  // archaic: bomen (trees)
+  braamstoelen: "/\u02C8bra\u02D0m\u02CCstu\u02D0l\u0259n/",
+  // blackberry bushes
+  bureautje: "/by\u02C8ro\u02D0tj\u0259/",
+  burgery: "/b\u028Fr\u0263\u0259\u02C8r\u025Bi/",
+  // archaic: burgerij
+  dagelix: "/\u02C8da\u02D0\u0263\u0259l\u026Aks/",
+  // archaic: dagelijks
+  dagschemer: "/\u02C8d\u0251\u0263\u02CCsxe\u02D0m\u0259r/",
+  dengenen: "/d\u025Bn\u02C8\u0263e\u02D0n\u0259n/",
+  // archaic: those
+  dese: "/\u02C8de\u02D0z\u0259/",
+  // archaic: deze
+  deselve: "/d\u0259\u02C8z\u025Blv\u0259/",
+  // archaic: dezelfde
+  dien: "/di\u02D0n/",
+  // archaic dative: that
+  draaide: "/\u02C8dra\u02D0id\u0259/",
+  duitsen: "/\u02C8d\u0153yts\u0259n/",
+  // archaic: German
+  eenvouds: "/\u02C8e\u02D0nv\u0251uts/",
+  // archaic: simplicity
+  eert: "/e\u02D0rt/",
+  egters: "/\u02C8\u025B\u0263t\u0259rs/",
+  // proper name
+  engelen: "/\u02C8\u025B\u014B\u0259l\u0259n/",
+  erbarremt: "/\u025Br\u02C8b\u0251r\u0259mt/",
+  // archaic: have mercy
+  erembodegem: "/\u02C8e\u02D0r\u0259m\u02CCbo\u02D0d\u025B\u0263\u0259m/",
+  erkers: "/\u02C8\u025Brk\u0259rs/",
+  flauwe: "/\u02C8fl\u0251u\u0259/",
+  frits: "/fr\u026Ats/",
+  gekend: "/\u0263\u0259\u02C8k\u025Bnt/",
+  // past participle: known
+  gekomen: "/\u0263\u0259\u02C8ko\u02D0m\u0259n/",
+  gekund: "/\u0263\u0259\u02C8k\u028Fnt/",
+  gemeenschappelijks: "/\u0263\u0259\u02C8me\u02D0nsx\u0251p\u0259l\u0259ks/",
+  // archaic genitive
+  geschrey: "/\u0263\u0259\u02C8sxr\u025Bi/",
+  // archaic: outcry
+  gestelt: "/\u0263\u0259\u02C8st\u025Blt/",
+  // archaic: placed
+  geverfd: "/\u0263\u0259\u02C8v\u025Brft/",
+  // painted, dyed
+  gewone: "/\u0263\u0259\u02C8wo\u02D0n\u0259/",
+  ghewelt: "/\u0263\u0259\u02C8w\u025Blt/",
+  // archaic: violence
+  godt: "/\u0263\u0254t/",
+  // archaic: God
+  goedkope: "/\u02C8\u0263utk\u02B0o\u02D0p\u0259/",
+  graaft: "/\u0263ra\u02D0ft/",
+  groeiden: "/\u02C8\u0263ru\u02D0id\u0259n/",
+  grooten: "/\u02C8\u0263ro\u02D0t\u0259n/",
+  // archaic: great
+  Haarlem: "/\u02C8ha\u02D0rl\u0259m/",
+  // city name
+  had: "/h\u0251t/",
+  hadden: "/\u02C8h\u0251d\u0259n/",
+  hare: "/\u02C8ha\u02D0r\u0259/",
+  // archaic: her
+  hele: "/\u02C8he\u02D0l\u0259/",
+  hemelsche: "/\u02C8he\u02D0m\u0259lsx\u0259/",
+  // archaic: heavenly
+  henri: "/\u0251\u0303\u02C8ri/",
+  // French name
+  herinner: "/h\u025B\u02C8r\u026An\u0259r/",
+  hispanje: "/h\u026As\u02C8p\u0251\u0272\u0259/",
+  // archaic: Spain
+  hooft: "/ho\u02D0ft/",
+  // archaic: hoofd
+  hooren: "/\u02C8ho\u02D0r\u0259n/",
+  // archaic: horen
+  immense: "/\u026A\u02C8m\u025Bns\u0259/",
+  inni: "/\u02C8\u026Ani/",
+  // proper name (Mulisch)
+  kapellekensbaan: "/ka\u02C8p\u025Bl\u0259k\u0259ns\u02CCba\u02D0n/",
+  kennelick: "/\u02C8k\u025Bn\u0259l\u0259k/",
+  // archaic: evident
+  kraanwagentje: "/\u02C8kra\u02D0n\u02CCwa\u02D0\u0263\u0259ntj\u0259/",
+  laatsten: "/\u02C8la\u02D0tst\u0259n/",
+  lande: "/\u02C8l\u0251nd\u0259/",
+  // archaic dative: land
+  lauriergracht: "/l\u0251u\u02C8ri\u02D0r\u0263r\u0251xt/",
+  lesen: "/\u02C8le\u02D0z\u0259n/",
+  // archaic: lezen
+  lesten: "/\u02C8l\u025Bst\u0259n/",
+  // archaic: last
+  louterende: "/\u02C8l\u0251ut\u0259r\u025Bnd\u0259/",
+  mooiste: "/\u02C8mo\u02D0ist\u0259/",
+  // superlative: prettiest
+  my: "/m\u025Bi/",
+  // archaic: mij
+  nassouwe: "/n\u0251\u02C8s\u0251u\u0259/",
+  // archaic: Nassau
+  niemandsbos: "/\u02C8ni\u02D0m\u0251nts\u02CCb\u0254s/",
+  // nobody's forest
+  oeroeg: "/\u02C8uru\u0263/",
+  // proper name (Haasse)
+  ondersaten: "/\u02C8\u0254nd\u0259r\u02CCza\u02D0t\u0259n/",
+  // archaic: subjects
+  ontwaakte: "/\u0254nt\u02C8wa\u02D0kt\u0259/",
+  onverveerd: "/\u02CC\u0254nv\u0259r\u02C8ve\u02D0rt/",
+  opgerezen: "/\u02C8\u0254p\u0263\u0259\u02CCre\u02D0z\u0259n/",
+  osewoudt: "/\u02C8o\u02D0z\u0259\u02CCw\u0251ut/",
+  // proper name (Hermans)
+  ouders: "/\u02C8\u0251ud\u0259rs/",
+  philips: "/\u02C8fil\u026Aps/",
+  plachten: "/\u02C8pl\u0251xt\u0259n/",
+  // archaic: used to
+  pleegde: "/\u02C8ple\u02D0\u0263d\u0259/",
+  po\u00EBtische: "/po\u02C8e\u02D0t\u026Asx\u0259/",
+  prince: "/\u02C8pr\u026Ans\u0259/",
+  // archaic: prins
+  prinse: "/\u02C8pr\u026Ans\u0259/",
+  // archaic variant
+  reed: "/re\u02D0t/",
+  romans: "/ro\u02C8m\u0251ns/",
+  saluyt: "/sa\u02D0\u02C8l\u0153yt/",
+  // archaic: greeting
+  sarphatistraat: "/s\u0251r\u02C8fa\u02D0tistra\u02D0t/",
+  // street name (Nescio)
+  scherpste: "/\u02C8sx\u025Brpst\u0259/",
+  schilderskade: "/\u02C8sx\u026Ald\u0259rs\u02CCka\u02D0d\u0259/",
+  // street name
+  Seynaeve: "/\u02C8s\u025Bina\u02D0v\u0259/",
+  // Belgian proper name
+  sien: "/si\u02D0n/",
+  // archaic: zien
+  smalle: "/\u02C8sm\u0251l\u0259/",
+  // inflected: narrow
+  spoorwegzate: "/\u02C8spo\u02D0r\u02CCw\u025Bxza\u02D0t\u0259/",
+  // archaic: railway junction
+  Steenwijk: "/\u02C8ste\u02D0n\u028B\u025Bik/",
+  // place name
+  stonden: "/\u02C8st\u0254nd\u0259n/",
+  stopte: "/\u02C8st\u0254pt\u0259/",
+  tamarindeboomen: "/ta\u02D0ma\u02D0\u02C8r\u026And\u0259\u02CCbo\u02D0m\u0259n/",
+  // archaic: tamarind trees
+  tegenwoordighe: "/\u02C8te\u02D0\u0263\u0259n\u02CCwo\u02D0rd\u0259\u0263\u0259/",
+  // archaic
+  terechtgekomen: "/t\u0259\u02C8r\u025Bxt\u0263\u0259\u02CCko\u02D0m\u0259n/",
+  termurenlaan: "/t\u025Br\u02C8my\u02D0r\u0259n\u02CCla\u02D0n/",
+  terugdenk: "/t\u0259\u02C8r\u028Fxd\u025B\u014Bk/",
+  // compound: think back
+  torentjes: "/\u02C8to\u02D0r\u0259ntj\u0259s/",
+  toverplaatjes: "/\u02C8to\u02D0v\u0259r\u02CCpla\u02D0tj\u0259s/",
+  // magic pictures
+  tracht: "/tr\u0251xt/",
+  // verb: try
+  tragische: "/\u02C8tra\u02D0\u0263\u026Asx\u0259/",
+  tusschen: "/\u02C8t\u028Fsx\u0259n/",
+  // archaic: tussen (between)
+  velden: "/\u02C8v\u025Bld\u0259n/",
+  vermoordde: "/v\u0259r\u02C8mo\u02D0rd\u0259/",
+  verschijnt: "/v\u0259r\u02C8sx\u025Bint/",
+  // appears
+  vlamde: "/\u02C8vl\u0251md\u0259/",
+  vloog: "/vlo\u02D0x/",
+  volcx: "/v\u0254lks/",
+  // archaic: volks
+  volle: "/\u02C8v\u0254l\u0259/",
+  voorschoten: "/\u02C8vo\u02D0rsxo\u02D0t\u0259n/",
+  // place name
+  voorzomermiddag: "/\u02C8vo\u02D0r\u02CCzo\u02D0m\u0259r\u02CCm\u026Ad\u0251x/",
+  vroege: "/\u02C8vru\u0263\u0259/",
+  warme: "/\u02C8v\u0251rm\u0259/",
+  weerd: "/we\u02D0rt/",
+  // archaic: waard
+  weggegaan: "/\u02C8v\u025Bx\u0263\u0259\u02CC\u0263a\u02D0n/",
+  werd: "/w\u025Brt/",
+  werkmenschen: "/\u02C8w\u025Brk\u02CCm\u025Bnsx\u0259n/",
+  // archaic: working people
+  "west-java": "/\u02CCw\u025Bst\u02C8ja\u02D0va/",
+  wintrop: "/\u02C8v\u026Antr\u0254p/",
+  // proper name (Mulisch)
+  wonderlijker: "/\u02C8\u028B\u0254nd\u0259rl\u025Bik\u0259r/",
+  // comparative: more wonderful
+  woonde: "/\u02C8wo\u02D0nd\u0259/",
+  wordt: "/w\u0254rt/",
+  yegelick: "/\u02C8je\u02D0\u0263\u0259l\u026Ak/",
+  // archaic: iedereen
+  zakkende: "/\u02C8z\u0251k\u0259nd\u0259/",
+  // sinking
+  zekere: "/\u02C8ze\u02D0k\u0259r\u0259/",
+  // a certain
+  zijne: "/\u02C8z\u025Bin\u0259/",
+  // archaic: zijn
+  zocht: "/z\u0254xt/",
+  zuivere: "/\u02C8z\u0153yv\u0259r\u0259/",
+  // pure (inflected)
+  zulke: "/\u02C8z\u028Flk\u0259/"
+};
+
+// src/overrides/or.ts
+var or_ = {
+  // Bande Utkala Janani (Odia anthem)
+  \u0B05\u0B19\u0B4D\u0B17\u0B47: "\u0254\u014B\u0261e",
+  // Jagannath Das — Odia Bhagabata
+  \u0B05\u0B28\u0B3E\u0B26\u0B3F: "\u0254na\u02D0d\u032Ai",
+  \u0B05\u0B2A\u0B4D\u0B30\u0B2E\u0B3F\u0B24: "\u0254p\u027E\u0254mit\u032A\u0254",
+  \u0B05\u0B30\u0B4D\u0B25\u0B47: "\u0254\u027Et\u032A\u02B0e",
+  \u0B05\u0B36\u0B47\u0B37: "\u0254\u0283e\u0282\u0254",
+  \u0B06\u0B17\u0B41\u0B01: "a\u02D0\u0261\u0169",
+  \u0B06\u0B24\u0B4D\u0B2E\u0B40\u0B5F: "a\u02D0t\u032Ami\u02D0j\u0254",
+  \u0B06\u0B30\u0B24: "a\u02D0\u027E\u0254t\u032A\u0254",
+  // Gangadhar Meher — Tapaswinee
+  \u0B06\u0B39\u0B4D\u0B32\u0B3E\u0B26: "a\u02D0hl\u032Aa\u02D0d\u032A\u0254",
+  \u0B07\u0B28\u0B4D\u0B26\u0B4D\u0B30\u0B28\u0B40\u0B33: "ind\u032A\u027E\u0254ni\u02D0l\u032A\u0254",
+  \u0B09\u0B24\u0B4D\u0B15\u0B33\u0B2D\u0B41\u0B2C\u0B28\u0B47: "ut\u032Ak\u0254l\u032A\u0254b\u02B1ub\u0254ne",
+  \u0B09\u0B24\u0B4D\u0B15\u0B33\u0B30: "ut\u032Ak\u0254l\u032A\u0254\u027E\u0254",
+  // Upendra Bhanja — Baidehisha Bilasa
+  \u0B09\u0B26\u0B4D\u0B2D\u0B2C: "ud\u032Ab\u02B1\u0254b\u0254",
+  \u0B0F\u0B23\u0B47: "e\u0273e",
+  \u0B14\u0B30\u0B38\u0B41: "\u0254w\u027E\u0254su",
+  // Rangabati (folk song)
+  \u0B15\u0B28\u0B15\u0B32\u0B24\u0B3E: "k\u0254n\u0254k\u0254l\u032A\u0254t\u032Aa\u02D0",
+  \u0B15\u0B33: "k\u0254l\u032A\u0254",
+  \u0B15\u0B39\u0B32\u0B4B: "k\u0254h\u0254l\u032Ao\u02D0",
+  \u0B15\u0B3E\u0B24\u0B30: "ka\u02D0t\u032A\u0254\u027E\u0254",
+  // Fakir Mohan Senapati — Chha Mana Atha Guntha
+  \u0B15\u0B3E\u0B30\u0B2C\u0B3E\u0B30: "ka\u02D0\u027E\u0254ba\u02D0\u027E\u0254",
+  \u0B15\u0B3E\u0B30\u0B3E\u0B26\u0B23\u0B4D\u0B21\u0B47: "ka\u02D0\u027Ea\u02D0d\u032A\u0254\u0273\u0256e",
+  \u0B15\u0B3F\u0B2E\u0B4D\u0B2A\u0B3E: "kimpa\u02D0",
+  \u0B15\u0B43\u0B37\u0B4D\u0B23\u0B2A\u0B3E\u0B26: "k\u027Eu\u0282\u0273\u0254pa\u02D0d\u032A\u0254",
+  \u0B15\u0B47\u0B2C\u0B3E: "keba\u02D0",
+  \u0B15\u0B47\u0B36\u0B3E: "ke\u0283a\u02D0",
+  \u0B15\u0B4B\u0B36: "ko\u02D0\u0283\u0254",
+  // Madhusudan Das — Utkala Santana
+  \u0B17\u0B19\u0B4D\u0B17\u0B3E: "\u0261\u0254\u014B\u0261a\u02D0",
+  \u0B17\u0B33\u0B41\u0B1B\u0B3F: "\u0261\u0254l\u032Aut\u0361\u0283\u02B0i",
+  \u0B17\u0B40\u0B24\u0B28\u0B3E\u0B26: "\u0261i\u02D0t\u032A\u0254na\u02D0d\u032A\u0254",
+  \u0B17\u0B41\u0B23\u0B3E\u0B33\u0B5F: "\u0261u\u0273a\u02D0l\u032A\u0254j\u0254",
+  \u0B17\u0B41\u0B01\u0B25\u0B3E: "\u0261\u0169t\u032A\u02B0a\u02D0",
+  \u0B17\u0B4B: "\u0261o\u02D0",
+  \u0B17\u0B4B\u0B26\u0B3E\u0B2C\u0B30\u0B40: "\u0261o\u02D0d\u032Aa\u02D0b\u0254\u027Ei\u02D0",
+  \u0B18\u0B1F\u0B47: "\u0261\u02B1\u0254\u0288e",
+  \u0B18\u0B41\u0B2E\u0B41\u0B38\u0B30\u0B05\u0B27\u0B3F\u0B2A: "\u0261\u02B1umus\u0254\u027E\u0254d\u032A\u02B1ip\u0254",
+  \u0B1A\u0B33\u0B47: "t\u0361\u0283\u0254l\u032Ae",
+  \u0B1A\u0B33\u0B47\u0B28\u0B3E\u0B39\u0B3F\u0B01: "t\u0361\u0283\u0254l\u032Aena\u02D0h\u0129",
+  \u0B1A\u0B3E\u0B30\u0B41: "t\u0361\u0283a\u02D0\u027Eu",
+  \u0B1A\u0B3E\u0B32\u0B3F\u0B2F\u0B3E\u0B06\u0B28\u0B4D\u0B24\u0B41: "t\u0361\u0283a\u02D0l\u032Aija\u02D0a\u02D0nt\u032Au",
+  \u0B1C\u0B17\u0B28\u0B4D\u0B28\u0B3E\u0B25\u0B47: "d\u0361\u0292\u0254\u0261\u0254nna\u02D0t\u032A\u02B0e",
+  \u0B1C\u0B28\u0B4D\u0B2E\u0B15\u0B3E\u0B33\u0B30\u0B41: "d\u0361\u0292\u0254nm\u0254ka\u02D0l\u032A\u0254\u027Eu",
+  \u0B1C\u0B2E\u0B3F\u0B26\u0B3E\u0B30: "d\u0361\u0292\u0254mid\u032Aa\u02D0\u027E\u0254",
+  \u0B1C\u0B3F\u0B24: "d\u0361\u0292it\u032A\u0254",
+  \u0B1C\u0B4D\u0B5F\u0B4B\u0B24\u0B3F\u0B30\u0B4D\u0B2E\u0B5F\u0B40: "d\u0361\u0292jo\u02D0t\u032Ai\u027Em\u0254ji\u02D0",
+  \u0B24\u0B26\u0B4D\u0B27\u0B24: "t\u032A\u0254d\u032Ad\u032A\u02B1\u0254t\u032A\u0254",
+  \u0B24\u0B28\u0B41\u0B15\u0B3E\u0B28\u0B4D\u0B24\u0B3F: "t\u032A\u0254nuka\u02D0nt\u032Ai",
+  \u0B24\u0B30\u0B19\u0B4D\u0B17\u0B47: "t\u032A\u0254\u027E\u0254\u014B\u0261e",
+  \u0B24\u0B30\u0B3F\u0B32\u0B47: "t\u032A\u0254\u027Eil\u032Ae",
+  \u0B24\u0B3E\u0B17\u0B47: "t\u032Aa\u02D0\u0261e",
+  \u0B24\u0B41\u0B39\u0B3F: "t\u032Auhi",
+  \u0B24\u0B4B\u0B39\u0B30: "t\u032Ao\u02D0h\u0254\u027E\u0254",
+  // Fakir Mohan Senapati — Odia Kahiki Daridra
+  \u0B26\u0B30\u0B3F\u0B26\u0B4D\u0B30\u0B24\u0B3E: "d\u032A\u0254\u027Eid\u032A\u027E\u0254t\u032Aa\u02D0",
+  \u0B26\u0B40\u0B30\u0B4D\u0B18\u0B3F\u0B15\u0B3E: "d\u032Ai\u02D0\u027E\u0261\u02B1ika\u02D0",
+  \u0B26\u0B40\u0B30\u0B4D\u0B18\u0B47: "d\u032Ai\u02D0\u027E\u0261\u02B1e",
+  \u0B26\u0B41\u0B03\u0B16: "d\u032Auk\u02B1\u0254",
+  \u0B26\u0B41\u0B30\u0B4D\u0B17\u0B24\u0B3F\u0B30: "d\u032Au\u027E\u0261\u0254t\u032Ai\u027E\u0254",
+  \u0B26\u0B47\u0B36\u0B2C\u0B3E\u0B38\u0B40: "d\u032Ae\u0283\u0254ba\u02D0si\u02D0",
+  \u0B26\u0B47\u0B36\u0B2E\u0B3E\u0B1F\u0B3F\u0B30\u0B47: "d\u032Ae\u0283\u0254ma\u02D0\u0288i\u027Ee",
+  \u0B26\u0B4D\u0B5F\u0B41\u0B24\u0B3F: "d\u032Ajut\u032Ai",
+  \u0B27\u0B28\u0B1E\u0B4D\u0B1C\u0B5F: "d\u032A\u02B1\u0254n\u0254\u0272d\u0361\u0292\u0254j\u0254",
+  \u0B27\u0B3E\u0B28\u0B30: "d\u032A\u02B1a\u02D0n\u0254\u027E\u0254",
+  \u0B28: "n\u0254",
+  \u0B28\u0B17\u0B26: "n\u0254\u0261\u0254d\u032A\u0254",
+  \u0B28\u0B28\u0B4D\u0B26\u0B28: "n\u0254nd\u032A\u0254n\u0254",
+  \u0B28\u0B2E\u0B07\u0B01: "n\u0254m\u0254\u0129",
+  \u0B28\u0B30\u0B4D\u0B15\u0B47: "n\u0254\u027Eke",
+  \u0B28\u0B3E\u0B30\u0B40\u0B19\u0B4D\u0B15: "na\u02D0\u027Ei\u02D0\u014Bk\u0254",
+  \u0B28\u0B3F\u0B30\u0B28\u0B4D\u0B24\u0B30\u0B47: "ni\u027E\u0254nt\u032A\u0254\u027Ee",
+  \u0B28\u0B3F\u0B39\u0B3F\u0B24: "nihit\u032A\u0254",
+  \u0B28\u0B40\u0B33\u0B3E\u0B2E\u0B4D\u0B2C\u0B41: "ni\u02D0l\u032Aa\u02D0mbu",
+  \u0B28\u0B43\u0B2A: "n\u027Eup\u0254",
+  \u0B2A\u0B21\u0B3C\u0B3F\u0B25\u0B3E\u0B09: "p\u0254\u0256\u032Ait\u032A\u02B0a\u02D0u",
+  \u0B2A\u0B26\u0B41: "p\u0254d\u032Au",
+  \u0B2A\u0B26\u0B47: "p\u0254d\u032Ae",
+  \u0B2A\u0B5F\u0B30\u0B47: "p\u0254j\u0254\u027Ee",
+  \u0B2A\u0B30\u0B3E: "p\u0254\u027Ea\u02D0",
+  \u0B2A\u0B42\u0B30\u0B4D\u0B2C\u0B2A\u0B41\u0B30\u0B41\u0B37: "pu\u02D0\u027Eb\u0254pu\u027Eu\u0282\u0254",
+  \u0B2A\u0B4B\u0B37\u0B23: "po\u02D0\u0282\u0254\u0273\u0254",
+  \u0B2A\u0B4D\u0B30\u0B1C\u0B4D\u0B1E\u0B3E: "p\u027E\u0254d\u0361\u0292\u0272a\u02D0",
+  \u0B2A\u0B4D\u0B30\u0B2C\u0B28\u0B4D\u0B27\u0B47: "p\u027E\u0254b\u0254nd\u032A\u02B1e",
+  \u0B2A\u0B4D\u0B30\u0B3E\u0B23\u0B40\u0B2E\u0B3E\u0B28\u0B19\u0B4D\u0B15: "p\u027Ea\u02D0\u0273i\u02D0ma\u02D0n\u0254\u014Bk\u0254",
+  \u0B2C\u0B28\u0B2D\u0B42\u0B2E\u0B3F: "b\u0254n\u0254b\u02B1u\u02D0mi",
+  \u0B2C\u0B28\u0B4D\u0B26\u0B28: "b\u0254nd\u032A\u0254n\u0254",
+  \u0B2C\u0B28\u0B4D\u0B26\u0B47: "b\u0254nd\u032Ae",
+  \u0B2C\u0B30\u0B39\u0B3F\u0B2C\u0B02\u0B36\u0B47: "b\u0254\u027E\u0254hib\u0254\u014B\u0283e",
+  \u0B2C\u0B30\u0B4D\u0B23\u0B4D\u0B23\u0B07: "b\u0254\u027E\u0273\u0273\u0254i",
+  \u0B2C\u0B3F\u0B1A\u0B3E\u0B30\u0B47: "bit\u0361\u0283a\u02D0\u027Ee",
+  \u0B2C\u0B3F\u0B28\u0B41: "binu",
+  \u0B2C\u0B3F\u0B2C\u0B47\u0B15: "bibek",
+  \u0B2C\u0B3F\u0B33\u0B3E\u0B38: "bil\u032Aa\u02D0s\u0254",
+  \u0B2C\u0B3F\u0B36\u0B3F\u0B37\u0B4D\u0B1F\u0B47: "bi\u0283i\u0282\u0288e",
+  \u0B2C\u0B3F\u0B39\u0B19\u0B4D\u0B17\u0B47: "bih\u0254\u014B\u0261e",
+  \u0B2C\u0B47\u0B28\u0B3F: "beni",
+  \u0B2C\u0B47\u0B36\u0B3E: "be\u0283a\u02D0",
+  \u0B2C\u0B4B\u0B32\u0B3E\u0B09: "bo\u02D0l\u032Aa\u02D0u",
+  \u0B2C\u0B4D\u0B5F\u0B25\u0B3E: "bj\u0254t\u032A\u02B0a\u02D0",
+  \u0B2C\u0B4D\u0B30\u0B1C: "b\u027E\u0254d\u0361\u0292\u0254",
+  \u0B2D\u0B23\u0B4D\u0B21\u0B3E\u0B30: "b\u02B1\u0254\u0273\u0256a\u02D0\u027E\u0254",
+  \u0B2D\u0B3E\u0B24\u0B43\u0B2D\u0B3E\u0B2C: "b\u02B1a\u02D0t\u032A\u027Eub\u02B1a\u02D0b\u0254",
+  \u0B2D\u0B3E\u0B37\u0B2E\u0B5F\u0B40: "b\u02B1a\u02D0\u0282\u0254m\u0254ji\u02D0",
+  \u0B2D\u0B40\u0B30\u0B41: "b\u02B1i\u02D0\u027Eu",
+  \u0B2D\u0B42\u0B27\u0B30\u0B2E\u0B3E\u0B33\u0B3E: "b\u02B1u\u02D0d\u032A\u02B1\u0254\u027E\u0254ma\u02D0l\u032Aa\u02D0",
+  \u0B2D\u0B47\u0B26\u0B3F: "b\u02B1ed\u032Ai",
+  \u0B2E\u0B15\u0B30\u0B28\u0B4D\u0B26: "m\u0254k\u0254\u027E\u0254nd\u032A\u0254",
+  \u0B2E\u0B19\u0B4D\u0B17\u0B30\u0B3E\u0B1C: "m\u0254\u014B\u0261\u0254\u027Ea\u02D0d\u0361\u0292\u0254",
+  \u0B2E\u0B28\u0B30\u0B47: "m\u0254n\u0254\u027Ee",
+  \u0B2E\u0B28\u0B4B\u0B39\u0B30: "m\u0254no\u02D0h\u0254\u027E\u0254",
+  \u0B2E\u0B2B\u0B38\u0B32\u0B30: "m\u0254p\u02B0\u0254s\u0254l\u032A\u0254\u027E\u0254",
+  \u0B2E\u0B30\u0B3E\u0B33\u0B2E\u0B3E\u0B33\u0B3F\u0B28\u0B40: "m\u0254\u027Ea\u02D0l\u032A\u0254ma\u02D0l\u032Aini\u02D0",
+  \u0B2E\u0B30\u0B4D\u0B2F\u0B4D\u0B5F\u0B3E\u0B26\u0B3E: "m\u0254\u027Ed\u0292ja\u02D0d\u032Aa\u02D0",
+  \u0B2E\u0B39\u0B3E\u0B1C\u0B28: "m\u0254ha\u02D0d\u0361\u0292\u0254n\u0254",
+  \u0B2E\u0B39\u0B3E\u0B1C\u0B28\u0B40: "m\u0254ha\u02D0d\u0361\u0292\u0254ni\u02D0",
+  \u0B2E\u0B3E\u0B32\u0B47: "ma\u02D0l\u032Ae",
+  \u0B2E\u0B3F\u0B36\u0B41: "mi\u0283u",
+  \u0B2E\u0B41\u0B16\u0B30\u0B3F\u0B24: "muk\u02B1\u0254\u027Eit\u032A\u0254",
+  \u0B30\u0B19\u0B4D\u0B17\u0B2C\u0B24\u0B40: "\u027E\u0254\u014B\u0261\u0254b\u0254t\u032Ai\u02D0",
+  \u0B30\u0B1C\u0B3E: "\u027E\u0254d\u0361\u0292a\u02D0",
+  \u0B30\u0B3E\u0B1C\u0B3F\u0B24: "\u027Ea\u02D0d\u0361\u0292it\u032A\u0254",
+  \u0B30\u0B3E\u0B28\u0B3F: "\u027Ea\u02D0ni",
+  \u0B30\u0B4B\u0B26\u0B28: "\u027Eo\u02D0d\u032A\u0254n\u0254",
+  \u0B36\u0B41\u0B23\u0B3E\u0B2F\u0B3E\u0B0F: "\u0283u\u0273a\u02D0ja\u02D0e",
+  \u0B36\u0B41\u0B26\u0B4D\u0B27: "\u0283ud\u032Ad\u032A\u02B1\u0254",
+  \u0B36\u0B41\u0B2D: "\u0283ub\u02B1\u0254",
+  \u0B36\u0B41\u0B2D\u0B4D\u0B30: "\u0283ub\u02B1\u027E\u0254",
+  \u0B36\u0B4B\u0B2D\u0B3E\u0B30: "\u0283ob\u02B1a\u02D0\u027E\u0254",
+  \u0B38\u0B2A\u0B28: "s\u0254p\u0254n\u0254",
+  \u0B38\u0B39\u0B41: "s\u0254hu",
+  \u0B38\u0B3E\u0B1C\u0B47: "sa\u02D0d\u0361\u0292e",
+  \u0B38\u0B41\u0B24\u0B3E: "sut\u032Aa\u02D0",
+  \u0B38\u0B41\u0B2A\u0B4D\u0B30\u0B15\u0B3E\u0B36: "sup\u027E\u0254ka\u02D0\u0283\u0254",
+  \u0B38\u0B47\u0B2E\u0B3E\u0B28\u0B19\u0B4D\u0B15\u0B20\u0B3E\u0B30\u0B47: "sema\u02D0n\u0254\u014Bk\u0254\u0288\u02B0a\u02D0\u027Ee",
+  \u0B38\u0B4B\u0B26\u0B30: "so\u02D0d\u032A\u0254\u027E\u0254",
+  \u0B38\u0B4D\u0B2B\u0B41\u0B30\u0B4D\u0B24\u0B4D\u0B24\u0B3F: "sp\u02B0u\u027Et\u032At\u032Ai",
+  "\u0B38\u0B4D\u0B71\u0B3E\u0B27\u0B40\u0B28": "swa\u02D0d\u032A\u02B1i\u02D0n\u0254",
+  \u0B39\u0B3E\u0B38\u0B2E\u0B5F\u0B40: "ha\u02D0s\u0254m\u0254ji\u02D0",
+  \u0B39\u0B3F\u0B24\u0B47: "hit\u032Ae",
+  \u0B39\u0B43\u0B26\u0B5F\u0B30\u0B47: "h\u027Eud\u032A\u0254j\u0254\u027Ee",
+  \u0B39\u0B43\u0B26\u0B47: "h\u027Eud\u032Ae",
+  \u0B39\u0B47\u0B32: "hel\u032A\u0254",
+  \u0B39\u0B47\u0B32\u0B3E\u0B24: "hel\u032Aa\u02D0t\u032A\u0254"
+};
+
+// src/overrides/pt.ts
+var pt = {
+  a: "/a/",
+  agora: "/a\u02C8\u0261o\u027Ea/",
+  ainda: "/a\u02C8\u0129da/",
+  alto: "/\u02C8awtu/",
+  alvejei: "/awve\u02C8\u0292ej/",
+  antes: "/\u02C8\xE3t\u0283is/",
+  // New sample overrides
+  Antigamente: "/\xE3t\u0283i\u0261a\u02C8m\u1EBDt\u0283i/",
+  aparecesse: "/apa\u027Ee\u02C8sesi/",
+  assembleia: "/as\u1EBD\u02C8bleja/",
+  autom\u00F3veis: "/awto\u02C8m\u0254vejs/",
+  autor: "/aw\u02C8to\u027E/",
+  aventura: "/av\u1EBD\u02C8tu\u027Ea/",
+  bater: "/ba\u02C8tex/",
+  beija: "/\u02C8bej\u0292a/",
+  bem: "/b\u1EBDj/",
+  calmo: "/\u02C8kawmu/",
+  cama: "/\u02C8k\xE3ma/",
+  certa: "/\u02C8s\u025B\u027Eta/",
+  chamam: "/\u02C8\u0283\xE3m\xE3w/",
+  como: "/\u02C8komu/",
+  contraparente: "/k\xF5t\u027Eapa\u02C8\u027E\u1EBDt\u0283i/",
+  costume: "/kos\u02C8t\u0169mi/",
+  cotovelo: "/koto\u02C8velu/",
+  cruzarmos: "/k\u027Euz\u02C8a\u027Emus/",
+  d: "/de/",
+  // abbreviation for Dom
+  de: "/d\u0292i/",
+  dei: "/dej/",
+  dele: "/\u02C8deli/",
+  dentro: "/\u02C8d\u1EBDt\u027Eu/",
+  deu: "/dew/",
+  dia: "/\u02C8d\u0292ia/",
+  direita: "/d\u0292i\u02C8\u027Eejta/",
+  direito: "/d\u0292i\u02C8\u027Eejtu/",
+  disco: "/\u02C8d\u0292isku/",
+  disposto: "/d\u0292is\u02C8postu/",
+  disse: "/\u02C8d\u0292isi/",
+  dois: "/dojs/",
+  \u00E9: "/\u025B/",
+  emprenhou: "/\u1EBDp\u027Ee\u02C8\u0272ow/",
+  encante: "/\u1EBD\u02C8k\xE3t\u0283i/",
+  entre: "/\u02C8\u1EBDt\u027Ei/",
+  escolherei: "/isko\u028Ee\u02C8\u027Eej/",
+  esf\u00EDngico: "/es\u02C8f\u0129\u0292iku/",
+  eterna: "/e\u02C8t\u025B\u027Ena/",
+  eu: "/ew/",
+  Europa: "/ew\u02C8\u027E\u0254pa/",
+  face: "/\u02C8fasi/",
+  faixas: "/\u02C8faj\u0283as/",
+  falsa: "/\u02C8fawsa/",
+  faz: "/fas/",
+  fim: "/f\u0129/",
+  Fita: "/\u02C8fita/",
+  fita: "/\u02C8fita/",
+  fitando: "/fi\u02C8t\xE3du/",
+  flameja: "/fla\u02C8me\u0292a/",
+  focinhando: "/fosi\u02C8\u0272\xE3du/",
+  foi: "/foj/",
+  fossem: "/\u02C8fos\u1EBDj/",
+  fosses: "/\u02C8fosis/",
+  fragorosas: "/f\u027Ea\u0261o\u02C8\u027Eozas/",
+  francisco: "/f\u027E\xE3\u02C8sisku/",
+  frente: "/\u02C8f\u027E\u1EBDt\u0283i/",
+  f\u00FAlgidos: "/\u02C8fuw\u0292idus/",
+  gente: "/\u02C8\u0292\u1EBDt\u0283i/",
+  girassol: "/\u0292i\u027Ea\u02C8s\u0254w/",
+  gosto: "/\u02C8\u0261ostu/",
+  grandes: "/\u02C8\u0261\u027E\xE3d\u0292is/",
+  havia: "/a\u02C8via/",
+  hei: "/ej/",
+  hesitei: "/ezi\u02C8tej/",
+  homens: "/\u02C8\xF5m\u1EBDjs/",
+  houve: "/\u02C8ovi/",
+  individuais: "/\u0129d\u0292ividu\u02C8ajs/",
+  instante: "/\u0129s\u02C8t\xE3t\u0283i/",
+  instituir: "/\u0129st\u0283itu\u02C8i\u027E/",
+  isso: "/\u02C8isu/",
+  jos\u00E9: "/\u0292o\u02C8z\u025B/",
+  liberdade: "/libe\u027E\u02C8dad\u0292i/",
+  lisboa: "/liz\u02C8boa/",
+  louca: "/\u02C8loka/",
+  louvor: "/lo\u02C8vo\u027E/",
+  lua: "/\u02C8lua/",
+  maior: "/ma\u02C8jo\u027E/",
+  mais: "/majs/",
+  mal: "/maw/",
+  margens: "/\u02C8ma\u027E\u0292\u1EBDjs/",
+  maus: "/maws/",
+  mesti\u00E7ara: "/mesti\u02C8sa\u027Ea/",
+  m\u00E9todo: "/\u02C8m\u025Btudu/",
+  mil: "/miw/",
+  mim: "/m\u0129/",
+  morder: "/mo\u027E\u02C8dex/",
+  morte: "/\u02C8m\u0254\u027Et\u0283i/",
+  morto: "/\u02C8mo\u027Etu/",
+  mostra: "/\u02C8m\u0254st\u027Ea/",
+  mulher: "/mu\u02C8\u028E\u025Bx/",
+  nacional: "/nasio\u02C8naw/",
+  nascimento: "/nasi\u02C8m\u1EBDtu/",
+  nele: "/\u02C8neli/",
+  nem: "/n\u1EBDj/",
+  nesse: "/\u02C8nesi/",
+  noite: "/\u02C8nojt\u0283i/",
+  nonada: "/no\u02C8nada/",
+  // nothing (Guimarães Rosa)
+  nosso: "/\u02C8nosu/",
+  novo: "/\u02C8novu/",
+  num: "/n\u0169/",
+  o: "/u/",
+  \u00F3: "/\u0254/",
+  ondas: "/\u02C8\xF5das/",
+  onde: "/\u02C8\xF5d\u0292i/",
+  ou: "/ow/",
+  ousadia: "/oza\u02C8d\u0292ia/",
+  outono: "/ow\u02C8tonu/",
+  outra: "/\u02C8owt\u027Ea/",
+  outro: "/\u02C8owt\u027Eu/",
+  para: "/\u02C8pa\u027Ea/",
+  parte: "/\u02C8pa\u027Et\u0283i/",
+  passadeira: "/pasa\u02C8dej\u027Ea/",
+  pode: "/\u02C8p\u0254d\u0292i/",
+  portugal: "/po\u027Etu\u02C8\u0261aw/",
+  posso: "/\u02C8posu/",
+  posta: "/\u02C8p\u0254sta/",
+  povo: "/\u02C8povu/",
+  praia: "/\u02C8p\u027Eaja/",
+  prometia: "/p\u027Eome\u02C8t\u0283ia/",
+  quanto: "/\u02C8kw\xE3tu/",
+  quarto: "/\u02C8kwa\u027Etu/",
+  quer: "/k\u025B\u027E/",
+  querer: "/ke\u02C8\u027Eex/",
+  ramalhete: "/xama\u02C8\u028Eet\u0283i/",
+  // mansion name (Eça de Queirós)
+  real: "/xe\u02C8aw/",
+  rebentavam: "/xeb\u1EBD\u02C8tav\xE3w/",
+  rei: "/xej/",
+  reino: "/\u02C8xejnu/",
+  remota: "/xe\u02C8m\u0254ta/",
+  retumbante: "/xet\u0169\u02C8b\xE3t\u0283i/",
+  rir: "/xi\u027E/",
+  riso: "/\u02C8xizu/",
+  rosto: "/\u02C8xostu/",
+  rua: "/\u02C8xua/",
+  s: "/s\xE3w/",
+  // abbreviation for São
+  saber: "/sa\u02C8bex/",
+  sal: "/saw/",
+  sei: "/sej/",
+  sem: "/s\u1EBDj/",
+  sempre: "/\u02C8s\u1EBDp\u027Ei/",
+  senhor: "/se\u02C8\u0272o\u027E/",
+  sequer: "/se\u02C8k\u025B\u027E/",
+  ser: "/\u02C8sex/",
+  sim: "/s\u0129/",
+  simples: "/\u02C8s\u0129plis/",
+  Sob: "/s\u0254b/",
+  sociais: "/sosi\u02C8ajs/",
+  sol: "/s\u0254w/",
+  sua: "/\u02C8sua/",
+  sublimaram: "/subli\u02C8ma\u027E\xE3w/",
+  tanto: "/\u02C8t\xE3tu/",
+  taprobana: "/tap\u027Eo\u02C8b\xE3na/",
+  // ancient Sri Lanka
+  te: "/t\u0283i/",
+  tempo: "/\u02C8t\u1EBDpu/",
+  ter: "/tex/",
+  tive: "/\u02C8t\u0283ivi/",
+  "toldam-lhe": "/\u02C8towd\xE3w\u028Ei/",
+  trapiche: "/t\u027Ea\u02C8pi\u0283i/",
+  triste: "/\u02C8t\u027Eist\u0283i/",
+  um: "/\u0169/",
+  uma: "/\u02C8\u0169ma/",
+  universo: "/uni\u02C8v\u025B\u027Esu/",
+  uso: "/\u02C8uzu/",
+  vale: "/\u02C8vali/",
+  verde: "/\u02C8ve\u027Ed\u0292i/",
+  verdes: "/\u02C8ve\u027Ed\u0292is/",
+  vez: "/ves/",
+  viagem: "/vi\u02C8a\u0292\u1EBDj/",
+  "viv\xEA-lo": "/vi\u02C8velu/",
+  "vou-me": "/\u02C8vomi/",
+  zelo: "/\u02C8zelu/",
+  zomba: "/\u02C8z\xF5ba/"
+};
+
+// src/overrides/ro.ts
+var ro = {
+  // New sample overrides
+  admirabil\u0103: "/admira\u02C8bil\u0259/",
+  alearg\u0103: "/a\u02C8le\u032Farg\u0259/",
+  alergi: "/a\u02C8lerd\u0292\u02B2/",
+  Armadia: "/ar\u02C8madja/",
+  b\u0103ie\u0219a\u0219: "/b\u0259je\u02C8\u0283a\u0283/",
+  "ca-ntr-un": "/kan\u02C8trun/",
+  C\u00E2rlibaba: "/k\u0268rli\u02C8baba/",
+  ciob\u0103nei: "/t\u0283ob\u0259\u02C8nej/",
+  ciople\u0219te: "/t\u0283o\u02C8ple\u0283te/",
+  cirezilor: "/t\u0283i\u02C8rezilor/",
+  cl\u0103di: "/kl\u0259\u02C8d\u02B2i/",
+  codrul: "/\u02C8kodrul/",
+  crengi: "/krend\u0292\u02B2/",
+  "croie\u0219te-\u021Bi": "/kro\u02C8je\u0283tet\u0283i/",
+  curat\u0103: "/ku\u02C8rat\u0259/",
+  deshame: "/des\u02C8hame/",
+  desprind: "/des\u02C8prind/",
+  "De\u0219teapt\u0103-te": "/de\u0283\u02C8te\u032Fapt\u0259te/",
+  fecioara: "/fet\u0283o\u02C8ara/",
+  ferestrelor: "/fe\u02C8restrelor/",
+  furtunoas\u0103: "/furtu\u02C8no\u032Fas\u0259/",
+  gem\u00E2nd: "/d\u0292e\u02C8m\u0268nd/",
+  grupurile: "/\u02C8\u0261rupurile/",
+  homer: "/\u02C8homer/",
+  humule\u0219tii: "/humu\u02C8le\u0283ti.i/",
+  \u00EEncepusem: "/\u0268nt\u0283e\u02C8pusem/",
+  \u00EEntinsese: "/\u0268ntin\u02C8sese/",
+  \u00EEntov\u0103r\u0103\u0219ind: "/\u0268ntov\u0259r\u0259\u02C8\u0283ind/",
+  intrat\u0103: "/in\u02C8trat\u0259/",
+  Jidovi\u021Ba: "/\u0292ido\u02C8vits\u0259/",
+  "l-a": "/la/",
+  "le-nt\xE2lnesc": "/lent\u0268l\u02C8nesk/",
+  luasem: "/lu\u02C8asem/",
+  lustruie\u0219te: "/lustru\u02C8je\u0283te/",
+  "m\xE2n\u0103-n": "/\u02C8m\u0268n\u0259n/",
+  "m\xE2ndr\u0103-n": "/\u02C8m\u0268ndr\u0259n/",
+  morminte: "/mor\u02C8minte/",
+  muceg\u0103it\u0103: "/mut\u0283e\u0261\u0259\u02C8it\u0259/",
+  na\u021Bional\u0103: "/natsio\u02C8nal\u0259/",
+  Obloanele: "/oblo\u032Fa\u02C8nele/",
+  "ochiu-adormit": "/\u02C8okju ador\u02C8mit/",
+  paraschi: "/para\u02C8ski/",
+  p\u0103rin\u021Bilor: "/p\u0259\u02C8rintsilor/",
+  "pe-ntinsele": "/pentin\u02C8sele/",
+  Prahovei: "/pra\u02C8hovej/",
+  r\u0103sar: "/r\u0259\u02C8sar/",
+  r\u0103zuie\u0219te: "/r\u0259zu\u02C8je\u0283te/",
+  r\u0103zvr\u0103tit\u0103: "/r\u0259zvr\u0259\u02C8tit\u0259/",
+  sc\u00E2r\u021B\u00E2iau: "/sk\u0268rts\u0268\u02C8jaw/",
+  sc\u00E2r\u021B\u00E2ie: "/sk\u0268r\u02C8ts\u0268je/",
+  scoal\u0103: "/\u02C8sko\u032Fal\u0259/",
+  scrumit: "/skru\u02C8mit/",
+  scuipat: "/skui\u02C8pat/",
+  "se-nchine": "/sen\u02C8kine/",
+  sfin\u021Bi: "/sfints\u02B2/",
+  sicriele: "/si\u02C8krijele/",
+  socoate: "/so\u02C8ko\u032Fate/",
+  Some\u0219ul: "/\u02C8some\u0283ul/",
+  \u0219oseaua: "/\u0283o\u02C8se\u032Fawa/",
+  stam: "/stam/",
+  strivesc: "/stri\u02C8vesk/",
+  sublocotenent: "/sublokote\u02C8nent/",
+  t\u0103ind: "/t\u0259\u02C8ind/",
+  tainele: "/\u02C8tajnele/",
+  "te-ad\xE2ncir\u0103": "/tead\u0268n\u02C8t\u0283ir\u0259/",
+  "toate-s": "/\u02C8to\u032Fates/",
+  tr\u0103snesc: "/tr\u0259s\u02C8nesk/",
+  ucid: "/u\u02C8t\u0283id/",
+  uneltele: "/u\u02C8neltele/",
+  v\u0103ii: "/\u02C8v\u0259ji/",
+  v\u00E2nturat\u0103: "/v\u0268ntu\u02C8rat\u0259/",
+  ve\u0219tm\u00E2nt: "/ve\u0283t\u02C8m\u0268nt/",
+  "vijelia-ngrozitoare": "/vi\u02C8\u0292eli.a \u014B\u0261rozi\u02C8to\u032Fare/",
+  "Vino-n": "/vi\u02C8non/"
+};
+
+// src/overrides/sv.ts
+var sv = {
+  \u00E4dlaste: "/\u02C8\u025B\u02D0dlast\u025B/",
+  \u00E4ngder: "/\u02C8\u025B\u014B\u02D0d\u025Br/",
+  // New entries for updated samples
+  anl\u00E4nt: "/\u02C8an\u02D0l\u025Bnt/",
+  arbetsrum: "/\u02C8ar\u02D0bets\u02CCr\u0275m/",
+  arvedel: "/\u02C8ar\u02D0ve\u02CCde\u02D0l/",
+  atenare: "/at\u025B\u02C8n\u0251\u02D0r\u025B/",
+  \u00E5ttiotv\u00E5: "/\u02C8\u0254t\u02D0\u026A\u0254\u02CCtvo\u02D0/",
+  \u00E5ttitalet: "/\u02C8\u0254t\u02D0\u026A\u02CCt\u0251\u02D0l\u025Bt/",
+  barnkullarnas: "/\u02C8b\u0251\u02D0\u0273\u02CCk\u0275l\u02D0arnas/",
+  // New sample overrides
+  Birck: "/b\u026Ark/",
+  b\u00F6nderna: "/\u02C8b\u0153nd\u025B\u0273a/",
+  brynja: "/\u02C8bryn\u02D0ja/",
+  bryt: "/bry\u02D0t/",
+  d\u00E4rn\u00E4st: "/\u02C8d\xE6\u02D0\u0273\u025Bst/",
+  dimmors: "/\u02C8d\u026Am\u02D0\u0254\u0282/",
+  eget: "/\u02C8e\u02D0\u0261\u025Bt/",
+  f\u00E5gels\u00E5ng: "/\u02C8fo\u02D0\u0261\u025Bl\u02CCs\u0254\u014B\u02D0/",
+  fj\u00E4llh\u00F6ga: "/\u02C8fj\u025Bl\u02D0\u02CCh\xF8\u02D0\u0261a/",
+  fjol\u00E5rets: "/\u02C8fju\u02D0l\u02CCo\u02D0r\u025Bts/",
+  fornstora: "/\u02C8f\u0254\u02D0\u0273\u02CCstu\u02D0ra/",
+  f\u00F6rt\u00E4tas: "/f\u0153r\u02C8t\u025B\u02D0tas/",
+  fredagskv\u00E4ll: "/\u02C8fre\u02D0da\u0261s\u02CCkv\u025Bl\u02D0/",
+  gamla: "/\u02C8\u0261am\u02D0la/",
+  gl\u00E4djerika: "/\u02C8\u0261l\u025Bd\u02D0j\u025B\u02CCri\u02D0ka/",
+  haga: "/\u02C8h\u0251\u02D0\u0261a/",
+  // place name
+  halvm\u00F6rkret: "/\u02C8halv\u02CCm\u0153rkr\u025Bt/",
+  "hedeby-\xF6n": "/\u02C8he\u02D0d\u025Bby\u02D0\u02CC\xF8\u02D0n/",
+  // place name
+  henrik: "/\u02C8h\u025Bnr\u026Ak/",
+  // proper name
+  h\u00F6llo: "/\u02C8h\xF8l\u02D0\u0254/",
+  improvisationer: "/\u026Ampr\u0254v\u026Asa\u02C8\u0267u\u02D0n\u025Br/",
+  Ingmarss\u00F6nerna: "/\u02C8\u026A\u014B\u02D0mar\u02CCs\xF8\u02D0n\u025B\u0273a/",
+  j\u00E4ttestaden: "/\u02C8j\u025Bt\u02D0\u025B\u02CCst\u0251\u02D0d\u025Bn/",
+  kriminalkommissarie: "/kr\u026Am\u026A\u02C8n\u0251\u02D0lk\u0254m\u026A\u02CCs\u0251\u02D0r\u026A\u025B/",
+  kristina: "/kr\u026A\u02C8sti\u02D0na/",
+  // proper name
+  kv\u00E4llsm\u00F6rkret: "/\u02C8kv\u025Bl\u02D0s\u02CCm\u0153rkr\u025Bt/",
+  l\u00E5ngstrump: "/\u02C8l\u0254\u014B\u02D0\u02CCstr\u0275mp/",
+  linh\u00E5rig: "/\u02C8li\u02D0n\u02CCho\u02D0r\u026A\u0261/",
+  ljusl\u00E5gor: "/\u02C8j\u0289\u02D0s\u02CClo\u02D0\u0261\u0254r/",
+  l\u00F6vsamlingar: "/\u02C8l\xF8\u02D0v\u02CCsam\u02D0l\u026A\u014B\u02D0ar/",
+  lyfte: "/\u02C8l\u028Fft\u025B/",
+  lyftes: "/\u02C8l\u028Fft\u025Bs/",
+  m\u00E4nniskor: "/\u02C8m\u025Bn\u02D0\u026A\u02CC\u0267\u0254r/",
+  m\u00E4nniskors: "/\u02C8m\u025Bn\u02D0\u026A\u02CC\u0267\u0254\u0282/",
+  Martin: "/\u02C8mar\u02D0t\u026An/",
+  m\u00E4ssfall: "/\u02C8m\u025Bs\u02D0\u02CCfal\u02D0/",
+  mosebacke: "/\u02C8mu\u02D0s\u025B\u02CCbak\u02D0\u025B/",
+  new: "/nju\u02D0/",
+  nilsson: "/\u02C8n\u026Al\u02D0s\u0254n/",
+  // proper name
+  nordamerika: "/\u02C8nu\u02D0\u0256a\u02CCme\u02D0r\u026Aka/",
+  numret: "/\u02C8n\u0275m\u02D0r\u025Bt/",
+  nytt: "/n\u028Ft\u02D0/",
+  odygd: "/\u02C8u\u02D0\u02CCdy\u0261d/",
+  ofruktsamt: "/\u02C8u\u02D0\u02CCfr\u0275ktsamt/",
+  \u00F6mt\u00E5ligare: "/\u02C8\xF8m\u02D0\u02CCto\u02D0l\u026A\u0261ar\u025B/",
+  \u00F6ppnad: "/\u02C8\u0153p\u02D0nad/",
+  orimligheter: "/u\u02D0\u02C8ri\u02D0ml\u026A\u0261\u02CChe\u02D0t\u025Br/",
+  oskar: "/\u02C8\u0254skar/",
+  // proper name
+  page: "/\u02C8p\u0251\u02D0\u0267\u0259/",
+  p\u00E4rontr\u00E4d: "/\u02C8p\xE6\u02D0r\u0254n\u02CCtr\u025B\u02D0d/",
+  paulun: "/pa\u02C8l\u0289\u02D0n/",
+  presentpappret: "/pr\u025B\u02C8s\u025Bnt\u02CCpap\u02D0r\u025Bt/",
+  proportionerad: "/pr\u0254p\u0254\u0282\u0254\u02C8ne\u02D0rad/",
+  roligast: "/\u02C8ru\u02D0l\u026A\u0261ast/",
+  r\u00F6tm\u00E5nadshetta: "/\u02C8r\xF8\u02D0t\u02CCmo\u02D0nads\u02CCh\u025Bt\u02D0a/",
+  runor: "/\u02C8r\u0289\u02D0n\u0254r/",
+  saffransblommorna: "/\u02C8saf\u02D0rans\u02CCbl\u0254m\u02D0\u0254\u0273a/",
+  scharlakanssol: "/\u02C8\u0267\u0251\u02D0rlakans\u02CCsu\u02D0l/",
+  si: "/si\u02D0/",
+  siljan: "/\u02C8s\u026Al\u02D0jan/",
+  silverne: "/\u02C8s\u026Alv\u025B\u0273\u025B/",
+  sk\u00E4ms: "/\u0267\u025Bm\u02D0s/",
+  sn\u00F6dropparna: "/\u02C8sn\xF8\u02D0\u02CCdr\u0254p\u02D0a\u0273a/",
+  sommarmorgon: "/\u02C8s\u0254mar\u02CCm\u0254r\u0261\u0254n/",
+  spiralgalax: "/sp\u026A\u02C8r\u0251\u02D0l\u0261a\u02CClaks/",
+  t\u00E4r: "/t\u025B\u02D0r/",
+  telefonsamtal: "/t\u025Bl\u025B\u02C8fo\u02D0n\u02CCsam\u02CCt\u0251\u02D0l/",
+  tillaga: "/t\u026Al\u02C8l\u0251\u02D0\u0261a/",
+  tjugosex: "/\u02C8\u0255\u0289\u02D0\u0261\u0254\u02CCs\u025Bks/",
+  upp\u00E5: "/\u02C8\u0275p\u02D0o\u02D0/",
+  uppgr\u00E4vda: "/\u02C8\u0275p\u02D0\u02CC\u0261r\u025B\u02D0vda/",
+  utflyttade: "/\u02C8\u0289\u02D0t\u02CCfl\u028Ft\u02D0ad\u025B/",
+  utsiktspunkt: "/\u02C8\u0289\u02D0t\u02CCs\u026Akts\u02CCp\u0275\u014Bkt/",
+  v\u00E4lvuxen: "/\u02C8v\u025B\u02D0l\u02CCv\u0275ks\u025Bn/",
+  vanger: "/\u02C8va\u014B\u02D0\u025Br/",
+  // surname
+  verklighetsgrund: "/\u02C8v\u025B\u02D0kl\u026A\u0261\u02CChe\u02D0ts\u02CC\u0261r\u0275nd/",
+  vilka: "/\u02C8v\u026Al\u02D0ka/",
+  vingad: "/\u02C8v\u026A\u014B\u02D0ad/",
+  voro: "/\u02C8vu\u02D0r\u0254/",
+  york: "/j\u0254rk/",
+  zephyrens: "/s\u025B\u02C8fy\u02D0r\u025Bns/"
+};
+
+// src/overrides/sw.ts
+var sw = {
+  // G2P handles most Swahili words. These overrides cover edge cases
+  // that G2P can't handle (Arabic loanwords, foreign names).
+  // Arabic loanword with "qi" — G2P doesn't have a rule for 'q'
+  sadiqi: "/sa\u02C8di\u02D0ki/",
+  ulithiqi: "/uli\u02C8\u03B8iki/",
+  // Arabic "kh" = /x/ (voiceless velar fricative), not /kh/
+  usiikhini: "/usii\u02C8xini/",
+  // Foreign name — 'c' alone is not a Swahili phoneme
+  Victoria: "/vikt\u0254\u02C8\u027Eia/"
+};
+
+// src/overrides/vi.ts
+var vi = {
+  chinh: "/t\u0255\u026A\u014B\u02E7/",
+  li\u00EAu: "/li\u0259w\u02E7/",
+  ng\u00E1c: "/\u014Bak\u02E7\u02E5/",
+  tru\u00E2n: "/\u0288w\u0259n\u02E7/"
+};
+
+// src/foreign.ts
+var IPA_SLASH_RE = /^\/|\/$/g;
+var WHITESPACE_SPLIT_RE = /(\s+)/;
+var WHITESPACE_RE = /^\s+$/;
+var LEADING_NON_LETTER_RE = /^[^\p{L}\p{M}]/u;
+var TRAILING_NON_LETTER_RE = /[^\p{L}\p{M}]$/u;
+var CONTRACTION_SPLIT_RE = /(?<=['-])|(?=['-])/;
+var khmerSegmenter = typeof Intl !== "undefined" && typeof Intl.Segmenter === "function" ? new Intl.Segmenter("km", { granularity: "word" }) : void 0;
+function segmentKhmerText(text) {
+  if (khmerSegmenter === void 0) {
+    return text;
+  }
+  const normalized = text.replaceAll("\u200B", " ");
+  const segments = [...khmerSegmenter.segment(normalized)];
+  let result = "";
+  for (let i = 0; i < segments.length; i++) {
+    const seg = segments[i];
+    result += seg.segment;
+    const next = segments[i + 1];
+    if (seg.isWordLike === true && next?.isWordLike === true) {
+      result += " ";
+    }
+  }
+  return result;
+}
+var LANGUAGES = [
+  { code: "ar", label: "Arabic" },
+  { code: "yue", label: "Cantonese" },
+  { code: "nl", label: "Dutch" },
+  { code: "eo", label: "Esperanto" },
+  { code: "fi", label: "Finnish" },
+  { code: "fr", label: "French" },
+  { code: "de", label: "German" },
+  { code: "is", label: "Icelandic" },
+  { code: "ja", label: "Japanese" },
+  { code: "km", label: "Khmer" },
+  { code: "ko", label: "Korean" },
+  { code: "ma", label: "Malay" },
+  { code: "zh", label: "Mandarin" },
+  { code: "nb", label: "Norwegian" },
+  { code: "or", label: "Odia" },
+  { code: "fa", label: "Persian" },
+  { code: "pt", label: "Portuguese" },
+  { code: "ro", label: "Romanian" },
+  { code: "es", label: "Spanish" },
+  { code: "sw", label: "Swahili" },
+  { code: "sv", label: "Swedish" },
+  { code: "vi", label: "Vietnamese" }
+];
+var IPA_WORD_OVERRIDES = {
+  ar,
+  de,
+  eo,
+  es,
+  fa,
+  fi,
+  fr,
+  is,
+  ja,
+  km,
+  ko,
+  ma,
+  nb,
+  nl,
+  or: or_,
+  pt,
+  ro,
+  sv,
+  sw,
+  vi
+};
+function ipaToIngglish(ipa) {
+  const clean = ipa.replaceAll(IPA_SLASH_RE, "").replaceAll(".", "");
+  const arpabet = ipaToArpabet(clean);
+  return (0, import_phonemes3.arpabetToIngglish)(arpabet);
+}
+function lookupIpa(dict, word) {
+  const { entries, lang } = dict;
+  const override = getIpaOverride(lang, word) ?? getIpaOverride(lang, word.toLowerCase());
+  if (override) {
+    return override;
+  }
+  const lower = word.toLowerCase();
+  const title = lower.charAt(0).toUpperCase() + lower.slice(1);
+  const stripped = (0, import_normalize.stripDiacritics)(lower);
+  if (entries[word] ?? entries[lower] ?? entries[title] ?? entries[stripped]) {
+    return entries[word] ?? entries[lower] ?? entries[title] ?? entries[stripped];
+  }
+  if (lower.includes("\xDF")) {
+    const ssLower = lower.replaceAll("\xDF", "ss");
+    const ssTitle = ssLower.charAt(0).toUpperCase() + ssLower.slice(1);
+    return entries[ssLower] ?? entries[ssTitle];
+  }
+  if (word.includes("'")) {
+    const curly = word.replaceAll("'", "\u2019");
+    const curlyLower = curly.toLowerCase();
+    const curlyResult = entries[curly] ?? entries[curlyLower];
+    if (curlyResult) {
+      return curlyResult;
+    }
+  }
+  if (Object.hasOwn(LEMMATIZERS, lang)) {
+    const lemmaResult = LEMMATIZERS[lang](entries, lower);
+    if (lemmaResult) {
+      return lemmaResult;
+    }
+  }
+  if (lang === "km") {
+    const compound = lookupKhmerCompound(entries, word);
+    if (compound !== void 0) {
+      return compound;
+    }
+  }
+  if (Object.hasOwn(G2P_CONVERTERS, lang)) {
+    return G2P_CONVERTERS[lang](lower);
+  }
+  return void 0;
+}
+var khmerMergedDict;
+var khmerDictKeys;
+function applyDefaultStress(arpabet) {
+  const hasStress = arpabet.some((p) => (0, import_phonemes3.isVowel)(p) && (0, import_phonemes3.getStress)(p) !== null);
+  if (hasStress) {
+    return arpabet;
+  }
+  const result = [...arpabet];
+  for (let i = result.length - 1; i >= 0; i--) {
+    if ((0, import_phonemes3.isVowel)(result[i])) {
+      result[i] = result[i] + "1";
+      break;
+    }
+  }
+  return result;
+}
+function decomposeKhmer(dict, keys, remaining, acc) {
+  if (remaining.length === 0) {
+    return acc;
+  }
+  for (const key of keys) {
+    const ipa = dict[key];
+    if (remaining.startsWith(key) && ipa !== void 0) {
+      const result = decomposeKhmer(dict, keys, remaining.slice(key.length), [...acc, ipa]);
+      if (result !== null) {
+        return result;
+      }
+    }
+  }
+  return null;
+}
+function getIpaOverride(lang, word) {
+  return IPA_WORD_OVERRIDES[lang]?.[word];
+}
+function ipaToFormat(ipa, format, lang) {
+  const clean = ipa.replaceAll(IPA_SLASH_RE, "").replaceAll(".", "");
+  const overrides = lang ? IPA_LANGUAGE_OVERRIDES[lang] : void 0;
+  const arpabet = applyDefaultStress(ipaToArpabet(clean, overrides));
+  return (0, import_phonemes3.arpabetToFormat)(arpabet, format, { disableRColoring: true });
+}
+function lookupKhmerCompound(entries, word) {
+  if (khmerMergedDict === void 0) {
+    khmerMergedDict = { ...entries };
+    const overrides = IPA_WORD_OVERRIDES.km;
+    if (overrides) {
+      for (const [k, v] of Object.entries(overrides)) {
+        khmerMergedDict[k] = v;
+      }
+    }
+  }
+  khmerDictKeys ?? (khmerDictKeys = Object.keys(khmerMergedDict).toSorted((a, b) => b.length - a.length));
+  const parts = decomposeKhmer(khmerMergedDict, khmerDictKeys, word, []);
+  if (parts === null || parts.length < 2) {
+    return void 0;
+  }
+  return parts.map((p) => p.replaceAll(IPA_SLASH_RE, "")).join(" ");
+}
+var NOT_FOUND_MARKER = "\uFFFD";
+var SENTENCE_END_RE = /[.!?。！？]$/;
+function translateForeign(text, dict, format = "ingglish") {
+  const tokens = translateForeignWithMapping(text, dict, format);
+  return tokens.map((t) => !t.matched && t.isWord ? NOT_FOUND_MARKER + t.original : t.translated).join("");
+}
+function translateForeignWithMapping(text, dict, format = "ingglish") {
+  const { lang } = dict;
+  let atSentenceStart = true;
+  const processed = lang === "km" ? segmentKhmerText(text) : text;
+  const tokens = [];
+  for (const segment of (0, import_normalize.normalizeApostrophes)(processed).split(WHITESPACE_SPLIT_RE)) {
+    if (WHITESPACE_RE.test(segment)) {
+      tokens.push({ isWord: false, matched: true, original: segment, translated: segment });
+      continue;
+    }
+    if (!segment) {
+      continue;
+    }
+    const leading = [];
+    const trailing = [];
+    let core = segment;
+    while (core.length > 0 && LEADING_NON_LETTER_RE.test(core)) {
+      leading.push(core[0]);
+      core = core.slice(1);
+    }
+    while (core.length > 0 && TRAILING_NON_LETTER_RE.test(core)) {
+      trailing.unshift(core.at(-1));
+      core = core.slice(0, -1);
+    }
+    if (!core) {
+      tokens.push({ isWord: false, matched: true, original: segment, translated: segment });
+      continue;
+    }
+    let casePattern = (0, import_normalize.detectCasePattern)(core);
+    const preservesCase = (0, import_phonemes3.getFormatPreservesCase)(format);
+    if (atSentenceStart && preservesCase && casePattern === "lower" && isCaselessWord(core)) {
+      casePattern = "capitalized";
+    }
+    atSentenceStart = SENTENCE_END_RE.test(trailing.join(""));
+    const leadStr = leading.join("");
+    const trailStr = trailing.join("");
+    const ipa = lookupIpa(dict, core);
+    if (ipa) {
+      const translated = ipaToFormat(ipa, format, lang);
+      const cased = preservesCase ? (0, import_normalize.applyCasePattern)(translated, casePattern) : translated;
+      tokens.push({
+        isWord: true,
+        matched: true,
+        original: segment,
+        translated: leadStr + cased + trailStr
+      });
+      continue;
+    }
+    const parts = core.split(CONTRACTION_SPLIT_RE);
+    if (parts.length > 1) {
+      const partIpas = parts.map((part, i) => {
+        if (part === "'" || part === "-") {
+          return;
+        }
+        let ipa2;
+        if (parts[i + 1] === "'") {
+          ipa2 = lookupIpa(dict, part + "'");
+        }
+        ipa2 ?? (ipa2 = lookupIpa(dict, part));
+        return ipa2;
+      });
+      const allFound = parts.every(
+        (part, i) => part === "'" || part === "-" || partIpas[i] !== void 0
+      );
+      if (allFound) {
+        const groups = [[]];
+        for (const [i, part_] of parts.entries()) {
+          const part = part_;
+          if (part === "'") {
+            continue;
+          }
+          if (part === "-") {
+            groups.push([]);
+            continue;
+          }
+          const ipa2 = partIpas[i];
+          groups.at(-1).push(ipa2.replaceAll(IPA_SLASH_RE, "").replaceAll(".", ""));
+        }
+        const translated2 = groups.map((ipas) => ipaToFormat(ipas.join(""), format, lang)).join("-");
+        const cased = preservesCase ? (0, import_normalize.applyCasePattern)(translated2, casePattern) : translated2;
+        tokens.push({
+          isWord: true,
+          matched: true,
+          original: segment,
+          translated: leadStr + cased + trailStr
+        });
+        continue;
+      }
+      let isFirstPart = true;
+      const translated = parts.map((part, i) => {
+        if (part === "'" || part === "-") {
+          return part;
+        }
+        const partCase = isFirstPart ? casePattern : (0, import_normalize.detectCasePattern)(part);
+        isFirstPart = false;
+        const partIpa = partIpas[i];
+        if (partIpa) {
+          const partTranslated = ipaToFormat(partIpa, format, lang);
+          return preservesCase ? (0, import_normalize.applyCasePattern)(partTranslated, partCase) : partTranslated;
+        }
+        return NOT_FOUND_MARKER + part;
+      });
+      if (translated.some(
+        (t, i) => parts[i] !== "'" && parts[i] !== "-" && !t.startsWith(NOT_FOUND_MARKER)
+      )) {
+        const translatedText = translated.map((t) => t.replaceAll(NOT_FOUND_MARKER, "")).join("");
+        tokens.push({
+          isWord: true,
+          matched: false,
+          original: segment,
+          translated: leadStr + translatedText + trailStr
+        });
+        continue;
+      }
+    }
+    tokens.push({ isWord: true, matched: false, original: segment, translated: segment });
+  }
+  return tokens;
+}
+function isCaselessWord(word) {
+  const ch = word[0];
+  return ch !== void 0 && ch.toUpperCase() === ch.toLowerCase();
+}
+
+// src/index.ts
+function registerIPA() {
+  (0, import_phonemes4.registerFormat)("ipa", {
+    forward: arpabetToIPARaw,
+    isLatinScript: true,
+    joinSeparator: " ",
+    label: "IPA",
+    preservesCase: false
+  });
+}
+// Annotate the CommonJS export names for ESM import in node:
+0 && (module.exports = {
+  LANGUAGES,
+  NOT_FOUND_MARKER,
+  arpabetPhonemeToIPA,
+  arpabetToIPARaw,
+  ipaToArpabetClean,
+  ipaToIngglish,
+  lookupIpa,
+  registerIPA,
+  segmentKhmerText,
+  translateForeign,
+  translateForeignWithMapping
+});