@ai-matrx/media 0.12.0 → 0.14.0

package/dist/voices.js

@@ -2314,13 +2314,498 @@ function voiceDisplayName(set, value) {
 function isLiveConversationVoice(id) {
   return LIVE_CONVERSATION_VOICES.some((voice) => voice.id === id);
 }
+
+// src/voices/speech-text.ts
+import {
+  findTableEnd,
+  isPipeLedRow,
+  replaceFences,
+  rowCells,
+  tableStartsAt,
+  unescapeCellPipes,
+  unwrapCodeSpans
+} from "@ai-matrx/content-ir/source";
+function speakTables(text) {
+  const lines = text.split("\n");
+  const out = [];
+  for (let i = 0; i < lines.length; i += 1) {
+    if (!tableStartsAt(lines, i)) {
+      out.push(lines[i] ?? "");
+      continue;
+    }
+    const end = findTableEnd(lines, i);
+    const headers = rowCells(lines[i] ?? "").map(unescapeCellPipes).filter((h) => h.length > 0);
+    const rowCount = lines.slice(i + 2, end).filter((line) => line.trim().length > 1).length;
+    const headerText = headers.length > 1 ? headers.slice(0, -1).join(", ") + ", and " + headers[headers.length - 1] : headers.length === 1 ? headers[0] : "unlabeled columns";
+    const rowText = rowCount === 1 ? "one row" : rowCount > 0 ? `${rowCount} rows` : "no rows";
+    out.push(`There is a table with ${rowText} of data provided for ${headerText}.`);
+    i = end - 1;
+  }
+  return out.join("\n");
+}
+var SPOKEN_CODE_LANGUAGES = /* @__PURE__ */ new Set([
+  "javascript",
+  "js",
+  "typescript",
+  "ts",
+  "python",
+  "py",
+  "java",
+  "csharp",
+  "cs",
+  "cpp",
+  "c++",
+  "c",
+  "go",
+  "rust",
+  "php",
+  "ruby",
+  "swift",
+  "kotlin",
+  "scala",
+  "sql",
+  "bash",
+  "shell",
+  "powershell",
+  "yaml",
+  "yml",
+  "json",
+  "xml",
+  "html",
+  "css",
+  "markdown",
+  "md"
+]);
+var COMMON_ABBREVIATION_EXPANSIONS = {
+  AI: "Artificial Intelligence",
+  API: "Application Programming Interface",
+  HTTP: "Hypertext Transfer Protocol",
+  HTTPS: "Hypertext Transfer Protocol Secure",
+  URL: "Uniform Resource Locator",
+  URI: "Uniform Resource Identifier",
+  JSON: "JavaScript Object Notation",
+  XML: "eXtensible Markup Language",
+  CSS: "Cascading Style Sheets",
+  HTML: "Hypertext Markup Language",
+  JS: "JavaScript",
+  TS: "TypeScript",
+  SQL: "Structured Query Language",
+  DB: "Database",
+  UI: "User Interface",
+  UX: "User Experience",
+  SEO: "Search Engine Optimization",
+  SDK: "Software Development Kit",
+  CLI: "Command Line Interface",
+  IDE: "Integrated Development Environment",
+  JWT: "JSON Web Token",
+  OAuth: "Open Authorization",
+  REST: "Representational State Transfer",
+  CRUD: "Create Read Update Delete",
+  MVC: "Model View Controller",
+  SPA: "Single Page Application",
+  SSR: "Server Side Rendering",
+  CSR: "Client Side Rendering",
+  PWA: "Progressive Web App",
+  DOM: "Document Object Model",
+  BOM: "Browser Object Model",
+  CDN: "Content Delivery Network",
+  CMS: "Content Management System",
+  ERP: "Enterprise Resource Planning",
+  CRM: "Customer Relationship Management",
+  SaaS: "Software as a Service",
+  PaaS: "Platform as a Service",
+  IaaS: "Infrastructure as a Service",
+  VPN: "Virtual Private Network",
+  LAN: "Local Area Network",
+  WAN: "Wide Area Network",
+  TCP: "Transmission Control Protocol",
+  UDP: "User Datagram Protocol",
+  IP: "Internet Protocol",
+  DNS: "Domain Name System",
+  FTP: "File Transfer Protocol",
+  SMTP: "Simple Mail Transfer Protocol",
+  POP3: "Post Office Protocol version 3",
+  IMAP: "Internet Message Access Protocol"
+};
+function escapeRegExp(s) {
+  return s.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
+}
+var COMMON_ABBREVIATION_BY_UPPERCASE = Object.fromEntries(
+  Object.keys(COMMON_ABBREVIATION_EXPANSIONS).map(
+    (abbreviation) => [abbreviation.toUpperCase(), abbreviation]
+  )
+);
+var COMMON_ABBREVIATION_PATTERN = new RegExp(
+  `\\b(${Object.keys(COMMON_ABBREVIATION_EXPANSIONS).sort((a, b) => b.length - a.length).map(escapeRegExp).join("|")})(['\u2019]s|s)?\\b`,
+  "gi"
+);
+var SPOKEN_AS_WORD = /* @__PURE__ */ new Set([
+  "JSON",
+  "OAuth",
+  "REST",
+  "CRUD",
+  "DOM",
+  "BOM",
+  "SaaS",
+  "PaaS",
+  "IaaS",
+  "LAN",
+  "WAN"
+]);
+var CASE_SENSITIVE_WORD_ACRONYMS = /* @__PURE__ */ new Set([
+  "REST",
+  "CRUD",
+  "SPA",
+  "DOM",
+  "BOM",
+  "LAN",
+  "WAN"
+]);
+function spokenFormForAbbreviation(match, sourceAbbreviation, suffix = "") {
+  const abbreviation = COMMON_ABBREVIATION_BY_UPPERCASE[sourceAbbreviation.toUpperCase()];
+  if (!abbreviation) return match;
+  if (CASE_SENSITIVE_WORD_ACRONYMS.has(abbreviation) && sourceAbbreviation !== abbreviation) {
+    return match;
+  }
+  if (SPOKEN_AS_WORD.has(abbreviation)) return `${abbreviation}${suffix}`;
+  const spoken = Array.from(abbreviation.toUpperCase()).join(" ");
+  return suffix ? `${spoken}'s` : spoken;
+}
+function normalizeSpeechAbbreviations(text) {
+  if (!text) return text;
+  return text.replace(COMMON_ABBREVIATION_PATTERN, spokenFormForAbbreviation);
+}
+var SPEECH_BLANK_WORD = "blank";
+function normalizeSpeechBlanks(text) {
+  if (!text || !text.includes("_")) return text;
+  const blank = /(?<![\p{L}\p{N}])_(?:[ \t]*_)+(?![\p{L}\p{N}])/gu;
+  return text.split("\n").map(
+    (line) => /^[ \t]*_[ \t_]*$/.test(line) ? line : line.replace(blank, ` ${SPEECH_BLANK_WORD} `)
+  ).join("\n");
+}
+function applyPronunciations(text, pairs) {
+  let out = text;
+  for (const { from, to } of pairs) {
+    const term = from.trim();
+    if (!term || !to) continue;
+    const re = new RegExp(
+      `(?<![\\p{L}\\p{N}])${escapeRegExp(term)}(?![\\p{L}\\p{N}])`,
+      "giu"
+    );
+    out = out.replace(re, to);
+  }
+  return out;
+}
+function parseMarkdownToText(markdown, options) {
+  if (!markdown || typeof markdown !== "string") return "";
+  const numberToWords = (num) => {
+    const ones = [
+      "zero",
+      "one",
+      "two",
+      "three",
+      "four",
+      "five",
+      "six",
+      "seven",
+      "eight",
+      "nine",
+      "ten",
+      "eleven",
+      "twelve",
+      "thirteen",
+      "fourteen",
+      "fifteen",
+      "sixteen",
+      "seventeen",
+      "eighteen",
+      "nineteen"
+    ];
+    const tens = [
+      "",
+      "",
+      "twenty",
+      "thirty",
+      "forty",
+      "fifty",
+      "sixty",
+      "seventy",
+      "eighty",
+      "ninety"
+    ];
+    const numInt = parseInt(num, 10);
+    if (numInt < 20) return ones[numInt] ?? num;
+    if (numInt < 100) {
+      const t = Math.floor(numInt / 10);
+      const o = numInt % 10;
+      return o === 0 ? tens[t] ?? num : `${tens[t] ?? ""}-${ones[o] ?? ""}`;
+    }
+    if (numInt < 1e3) {
+      const h = Math.floor(numInt / 100);
+      const remainder = numInt % 100;
+      return remainder === 0 ? `${ones[h]} hundred` : `${ones[h]} hundred ${numberToWords(String(remainder))}`;
+    }
+    return String(numInt);
+  };
+  const emojiMap = {
+    "\u{1F60A}": "smiling face",
+    "\u{1F602}": "laughing face",
+    "\u2764\uFE0F": "heart",
+    "\u{1F44D}": "thumbs up",
+    "\u{1F44E}": "thumbs down",
+    "\u{1F525}": "fire",
+    "\u2B50": "star",
+    "\u2705": "check mark",
+    "\u274C": "cross mark",
+    "\u26A0\uFE0F": "warning",
+    "\u{1F680}": "rocket",
+    "\u{1F4A1}": "light bulb",
+    "\u{1F3AF}": "target",
+    "\u{1F4F1}": "mobile phone",
+    "\u{1F4BB}": "laptop",
+    "\u{1F31F}": "glowing star",
+    "\u{1F517}": "link",
+    "\u{1F4E7}": "email",
+    "\u{1F4C4}": "document",
+    "\u{1F4CA}": "chart",
+    "\u{1F6E0}\uFE0F": "tools",
+    "\u2699\uFE0F": "gear",
+    "\u{1F527}": "wrench",
+    "\u{1F4C8}": "chart increasing",
+    "\u{1F4C9}": "chart decreasing",
+    "\u{1F389}": "party popper",
+    "\u{1F4AA}": "flexed biceps",
+    "\u{1F914}": "thinking face",
+    "\u{1F4AD}": "thought bubble",
+    "\u{1F440}": "eyes",
+    "\u{1F44B}": "waving hand",
+    "\u2728": "sparkles",
+    "\u{1F50D}": "magnifying mx-glass",
+    "\u{1F4DD}": "memo",
+    "\u{1F4CB}": "clipboard",
+    "\u{1F4C5}": "calendar",
+    "\u23F0": "alarm clock",
+    "\u{1F512}": "locked",
+    "\u{1F513}": "unlocked",
+    "\u{1F3C6}": "trophy",
+    "\u{1F396}\uFE0F": "military medal",
+    "\u{1F397}\uFE0F": "reminder ribbon",
+    "\u{1F3C5}": "sports medal",
+    "\u{1F947}": "first place medal",
+    "\u{1F948}": "second place medal",
+    "\u{1F949}": "third place medal"
+  };
+  const measurementUnits = {
+    // Weight
+    lbs: "pound",
+    // Changed from 'pounds'
+    lb: "pound",
+    oz: "ounce",
+    // Changed from 'ounces'
+    kg: "kilogram",
+    // Changed from 'kilograms'
+    gm: "gram",
+    // Changed from 'grams'
+    mg: "milligram",
+    // Changed from 'milligrams'
+    // Distance/Length
+    km: "kilometer",
+    // Changed from 'kilometers'
+    cm: "centimeter",
+    // Changed from 'centimeters'
+    mm: "millimeter",
+    // Changed from 'millimeters'
+    ft: "foot",
+    // Changed from 'feet' (Irregular plural)
+    mi: "mile",
+    // Changed from 'miles'
+    // Speed
+    mph: "mile per hour",
+    // Changed from 'miles per hour'
+    kph: "kilometer per hour",
+    // Changed from 'kilometers per hour'
+    kmh: "kilometer per hour",
+    // Changed from 'kilometers per hour'
+    // Volume
+    ml: "milliliter",
+    // Changed from 'milliliters'
+    mL: "milliliter",
+    // Changed from 'milliliters'
+    gal: "gallon",
+    // Changed from 'gallons'
+    qt: "quart",
+    // Changed from 'quarts'
+    tbsp: "tablespoon",
+    // Changed from 'tablespoons'
+    tsp: "teaspoon",
+    // Changed from 'teaspoons'
+    // Area
+    sqft: "square foot",
+    // Changed from 'square feet' (Irregular plural)
+    sqm: "square meter",
+    // Changed from 'square meters'
+    // Time
+    sec: "second",
+    // Changed from 'seconds'
+    min: "minute",
+    // Changed from 'minutes'
+    hr: "hour",
+    // Changed from 'hours'
+    hrs: "hour",
+    // Changed from 'hours'
+    ms: "millisecond",
+    // Changed from 'milliseconds'
+    // Other common units
+    psi: "pound per square inch",
+    // Changed from 'pounds per square inch'
+    rpm: "revolution per minute",
+    // Changed from 'revolutions per minute'
+    bpm: "beat per minute"
+    // Changed from 'beats per minute'
+  };
+  const stripReasoningTags = (input) => {
+    return input.replace(/<(thinking|reasoning|think|reason)\b[^>]*>[\s\S]*?<\/\1>/gi, "").replace(/<\/?(?:thinking|reasoning|think|reason)\b[^>]*>/gi, "");
+  };
+  const stripLeadingMarkdown = (input) => {
+    let text = input.replace(/^\uFEFF/, "").replace(/^[ \t]*\r?\n/, "");
+    text = text.replace(
+      /^[ \t]*(?:#{1,6}[ \t]+|>[ \t]*|[-*+][ \t]+|\d+\.[ \t]+|-[ \t]*\[[ xX]\][ \t]+)/,
+      ""
+    );
+    return text;
+  };
+  const source = normalizeSpeechBlanks(
+    options?.pronunciations?.length ? applyPronunciations(markdown, options.pronunciations) : markdown
+  );
+  let result = speakTables(unwrapCodeSpans(replaceFences(stripLeadingMarkdown(stripReasoningTags(source)), ({ lang }) => {
+    const language = lang.toLowerCase();
+    if (language === "mermaid") return "Please see the diagram provided.";
+    return SPOKEN_CODE_LANGUAGES.has(language) ? `Please see the ${language} code provided.` : "Please see the code provided.";
+  }))).split("\n").map((line) => isPipeLedRow(line) && line.trimEnd().endsWith("|") ? "" : line).join("\n").replace(
+    /\b(\d+)\s*[-–—]\s*(\d+)\b/g,
+    (_match, a, b) => `${numberToWords(a)} to ${numberToWords(b)}`
+  ).replace(
+    /^#{1,6}\s+(.+)$/gm,
+    (_m, t) => /[.!?:]$/.test(t.trim()) ? t : `${t.trim()}.`
+  ).replace(/\/\/\s*/g, "").replace(/--\s*/g, "").replace(/#\s+/g, "").replace(/;\s*/g, "").replace(/(\*\*|__)(.*?)\1/g, "$2").replace(/(\*|_)(.*?)\1/g, "$2").replace(/~~(.*?)~~/g, "$1").replace(/==(.*?)==/g, "$1").replace(/\[([^\]]+)\]\(([^)]+)\)/g, (_match, text, url) => {
+    if (url.startsWith("mailto:")) {
+      return `${text}. Email address provided.`;
+    }
+    const fileExtensions = [
+      ".pdf",
+      ".doc",
+      ".docx",
+      ".xls",
+      ".xlsx",
+      ".ppt",
+      ".pptx",
+      ".txt",
+      ".md",
+      ".json",
+      ".xml",
+      ".csv",
+      ".zip"
+    ];
+    const hasFileExtension = fileExtensions.some(
+      (ext) => url.toLowerCase().includes(ext)
+    );
+    if (hasFileExtension) {
+      return `${text}. Document link provided.`;
+    }
+    return `${text}. Link provided.`;
+  }).replace(
+    /!\[([^\]]*)\]\(([^)]+)\)/g,
+    ($0, alt) => alt ? `${alt}. Image provided.` : "Image provided."
+  ).replace(/^>\s*(.+)$/gm, "Quote: $1").replace(/^-\s*\[x\]\s+(.+)$/gm, "Completed task: $1").replace(/^-\s*\[\s*\]\s+(.+)$/gm, "Pending task: $1").replace(/^([-*+])\s+(.+)$/gm, "$2").replace(
+    /^(\d+)\.\s+(.+)$/gm,
+    (_match, num, content) => `Number ${numberToWords(num)}: ${content}`
+  ).replace(/\[\^(\d+)\]/g, "Reference $1").replace(/^\[\d+\]:\s*(.+)$/gm, "Reference: $1").replace(/^(\*{3,}|-{3,}|_{3,})$/gm, "").replace(/\$([^$]+)\$/g, "Mathematical expression: $1").replace(/\$\$([^$]+)\$\$/g, "Mathematical formula: $1").replace(
+    /(\+\d{1,3}[-.\s]?)?\(?\d{3}\)?[-.\s]?\d{3}[-.\s]?\d{4}/g,
+    "Phone number provided."
+  ).replace(
+    /([a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,})/g,
+    "Email address: $1"
+  ).replace(/\$(\d+(?:,\d{3})*(?:\.\d{2})?)/g, "$1 dollars").replace(/€(\d+(?:[.,]\d{3})*(?:[.,]\d{2})?)/g, "$1 euros").replace(/£(\d+(?:,\d{3})*(?:\.\d{2})?)/g, "$1 pounds").replace(/¥(\d+(?:,\d{3})*(?:\.\d{2})?)/g, "$1 yen").replace(
+    /(\d{4})[-/](\d{1,2})[-/](\d{1,2})/g,
+    (_match, year, month, day) => {
+      const monthNames = [
+        "January",
+        "February",
+        "March",
+        "April",
+        "May",
+        "June",
+        "July",
+        "August",
+        "September",
+        "October",
+        "November",
+        "December"
+      ];
+      const monthName = monthNames[parseInt(month) - 1] || month;
+      return `${monthName} ${day}, ${year}`;
+    }
+  ).replace(
+    /(\d{1,2}):(\d{2})\s*(AM|PM|am|pm)?/g,
+    (_match, hour, minute, period) => {
+      const hourNum = parseInt(hour);
+      const periodText = period ? period.toUpperCase() === "AM" ? "A.M." : "P.M." : "";
+      return `${hourNum} ${minute} ${periodText}`.trim();
+    }
+  ).replace(
+    /[\u{1F600}-\u{1F64F}]|[\u{1F300}-\u{1F5FF}]|[\u{1F680}-\u{1F6FF}]|[\u{1F1E0}-\u{1F1FF}]|[\u{2600}-\u{26FF}]|[\u{2700}-\u{27BF}]/gu,
+    (emoji) => {
+      return emojiMap[emoji] || "emoji";
+    }
+  ).replace(COMMON_ABBREVIATION_PATTERN, spokenFormForAbbreviation).replace(
+    /(\d+(?:\.\d+)?)\s*(lbs|lb|oz|kg|gm|mg|km|cm|mm|ft|mi|mph|kph|kmh|ml|mL|gal|qt|tbsp|tsp|sqft|sqm|sec|min|hr|hrs|ms|psi|rpm|bpm)\b/gi,
+    (match, number, unit) => {
+      const unitLower = unit.toLowerCase();
+      const unitText = measurementUnits[unitLower] || measurementUnits[unit] || unit;
+      return `${number} ${unitText}`;
+    }
+  ).replace(
+    /\b(lbs|lb|oz|kg|gm|mg|km|cm|mm|mph|kph|kmh|ml|mL|gal|qt|tbsp|tsp|sqft|sqm|psi|rpm|bpm)(?=\s|$|[.,;!?])/gi,
+    (match) => {
+      const unitLower = match.toLowerCase();
+      return measurementUnits[unitLower] || measurementUnits[match] || match;
+    }
+  ).replace(/©/g, "copyright").replace(/®/g, "registered trademark").replace(/™/g, "trademark").replace(/°/g, "degrees").replace(/±/g, "plus or minus").replace(/≈/g, "approximately").replace(/≠/g, "not equal to").replace(/≤/g, "less than or equal to").replace(/≥/g, "greater than or equal to").replace(/=/g, " equals ").replace(/→/g, "arrow").replace(/←/g, "left arrow").replace(/↑/g, "up arrow").replace(/↓/g, "down arrow").replace(/<\/?[^>]+(>|$)/g, "").replace(/([^\s.!?,;:\-])(\n)/g, "$1.\n").replace(/\s+/g, " ").trim();
+  return result;
+}
+
+// src/voices/read-aloud-languages.ts
+var READ_ALOUD_LANGUAGES = [
+  { code: "en", englishName: "English", nativeName: "English" },
+  { code: "es", englishName: "Spanish", nativeName: "Espa\xF1ol" },
+  { code: "fr", englishName: "French", nativeName: "Fran\xE7ais" },
+  { code: "de", englishName: "German", nativeName: "Deutsch" },
+  { code: "it", englishName: "Italian", nativeName: "Italiano" },
+  { code: "pt", englishName: "Portuguese", nativeName: "Portugu\xEAs" },
+  { code: "zh", englishName: "Chinese", nativeName: "\u4E2D\u6587" },
+  { code: "ja", englishName: "Japanese", nativeName: "\u65E5\u672C\u8A9E" },
+  { code: "ko", englishName: "Korean", nativeName: "\uD55C\uAD6D\uC5B4" },
+  { code: "ru", englishName: "Russian", nativeName: "\u0420\u0443\u0441\u0441\u043A\u0438\u0439" },
+  { code: "hi", englishName: "Hindi", nativeName: "\u0939\u093F\u0928\u094D\u0926\u0940" },
+  { code: "nl", englishName: "Dutch", nativeName: "Nederlands" },
+  { code: "pl", englishName: "Polish", nativeName: "Polski" },
+  { code: "sv", englishName: "Swedish", nativeName: "Svenska" },
+  { code: "tr", englishName: "Turkish", nativeName: "T\xFCrk\xE7e" },
+  { code: "fa", englishName: "Persian", nativeName: "\u0641\u0627\u0631\u0633\u06CC" }
+];
+function readAloudLanguage(code) {
+  return READ_ALOUD_LANGUAGES.find((l) => l.code === code) ?? READ_ALOUD_LANGUAGES[0];
+}
 export {
   ASSISTANT_VOICE_ID,
   CARTESIA_API_VERSION,
+  COMMON_ABBREVIATION_EXPANSIONS,
   LEGACY_DEFAULT_VOICE_ID,
   LIVE_CONVERSATION_SAMPLE_MODEL,
   LIVE_CONVERSATION_VOICES,
   READING_VOICE_ID,
+  READ_ALOUD_LANGUAGES,
+  SPEECH_BLANK_WORD,
   TTS_DEFAULT_SPEED,
   TTS_DEFAULT_VOLUME,
   TTS_MODEL_ID,
@@ -2331,6 +2816,10 @@ export {
   allVoices,
   availableVoices,
   isLiveConversationVoice,
+  normalizeSpeechAbbreviations,
+  normalizeSpeechBlanks,
+  parseMarkdownToText,
+  readAloudLanguage,
   resolveSpeed,
   resolveVoiceId,
   voiceDisplayName,