@leelaing/transcripts 0.1.0

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Lee Charles Laing
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,92 @@
+# @lcharleslaing/transcripts
+
+Transcript and song-lyric parsing utilities for SRT, VTT, LRC, TXT, JSON, and Whisper Music phrase JSON.
+
+## Supported formats
+
+- SRT
+- WebVTT / VTT
+- LRC
+- TXT
+- Generic JSON segment/chunk formats
+- Whisper Music phrase JSON, including `.phrase.json` and `.phrases.json`
+
+All formats normalize to the same structure:
+
+```ts
+interface TranscriptLine {
+  start: number | null;
+  end: number | null;
+  text: string;
+}
+
+interface Transcript {
+  format: TranscriptFormat;
+  lines: TranscriptLine[];
+  text: string;
+}
+```
+
+Phrase-level Whisper Music chunks stay phrase-level; they are not merged into larger subtitle blocks.
+
+## Usage
+
+```ts
+import { detectTranscriptFormat, parseTranscript } from '@lcharleslaing/transcripts';
+
+const content = '[00:01.20]First lyric line\n[00:03.40]Second lyric line';
+
+const format = detectTranscriptFormat(content, 'song.lrc');
+const transcript = parseTranscript(content, { filename: 'song.lrc' });
+
+console.log(format); // lrc
+console.log(transcript.lines);
+```
+
+You can force a known format:
+
+```ts
+const transcript = parseTranscript(jsonContent, {
+  format: 'phrase-json'
+});
+```
+
+## JSON compatibility
+
+The JSON parser recognizes common collections such as:
+
+- `segments`
+- `lines`
+- `items`
+- `transcript`
+- `phrases`
+- `phrase_segments`
+- `phraseSegments`
+- `chunks`
+
+Common text fields include `text`, `phrase`, `content`, `line`, `lyrics`, and `lyric`.
+
+Common timing fields include `start`, `start_time`, `startTime`, `end`, `end_time`, `endTime`, `duration`, and related aliases.
+
+## Development
+
+```bash
+npm install
+npm start
+```
+
+## Quality checks
+
+```bash
+npm run check
+```
+
+## Build
+
+```bash
+npm run build
+```
+
+## License
+
+MIT

package/dist/index.cjs

@@ -0,0 +1,189 @@
+'use strict';
+
+// src/index.ts
+function cleanText(value) {
+  return typeof value === "string" ? value.replace(/\s+/g, " ").trim() : "";
+}
+function toSeconds(value) {
+  if (typeof value === "number" && Number.isFinite(value)) return value;
+  if (typeof value !== "string") return null;
+  const normalized = value.trim().replace(",", ".");
+  if (!normalized) return null;
+  if (/^\d+(?:\.\d+)?$/.test(normalized)) {
+    const parsed = Number(normalized);
+    return Number.isFinite(parsed) ? parsed : null;
+  }
+  const parts = normalized.split(":").map(Number);
+  if (parts.some((part) => !Number.isFinite(part))) return null;
+  if (parts.length === 3) return parts[0] * 3600 + parts[1] * 60 + parts[2];
+  if (parts.length === 2) return parts[0] * 60 + parts[1];
+  return null;
+}
+function buildTranscript(format, lines) {
+  const filtered = lines.map((line) => ({ ...line, text: cleanText(line.text) })).filter((line) => line.text.length > 0);
+  return {
+    format,
+    lines: filtered,
+    text: filtered.map((line) => line.text).join("\n")
+  };
+}
+function parseTimedBlocks(content, format) {
+  const normalized = content.replace(/^\uFEFF/, "").replace(/\r/g, "");
+  const body = format === "vtt" ? normalized.replace(/^WEBVTT[^\n]*\n+/, "") : normalized;
+  const blocks = body.split(/\n{2,}/);
+  const lines = [];
+  for (const block of blocks) {
+    const rows = block.split("\n").filter(Boolean);
+    if (rows.length < 2) continue;
+    const timingIndex = rows.findIndex((row) => row.includes("-->"));
+    if (timingIndex < 0) continue;
+    const timing = rows[timingIndex].split("-->");
+    const start = toSeconds(timing[0]);
+    const end = toSeconds(timing[1]?.trim().split(/\s+/)[0]);
+    const text = rows.slice(timingIndex + 1).join(" ");
+    if (start !== null || end !== null || cleanText(text)) {
+      lines.push({ start, end, text });
+    }
+  }
+  return buildTranscript(format, lines);
+}
+function parseLrc(content) {
+  const lines = [];
+  for (const rawLine of content.replace(/\r/g, "").split("\n")) {
+    const matches = [...rawLine.matchAll(/\[(\d{1,3}):(\d{2}(?:\.\d{1,3})?)\]/g)];
+    const text = rawLine.replace(/\[[^\]]+\]/g, "").trim();
+    if (!text || matches.length === 0) continue;
+    for (const match of matches) {
+      const minutes = Number(match[1]);
+      const seconds = Number(match[2]);
+      lines.push({ start: minutes * 60 + seconds, end: null, text });
+    }
+  }
+  lines.sort((a, b) => (a.start ?? 0) - (b.start ?? 0));
+  for (let i = 0; i < lines.length - 1; i += 1) {
+    lines[i].end = lines[i + 1].start;
+  }
+  return buildTranscript("lrc", lines);
+}
+function parseTxt(content) {
+  return buildTranscript(
+    "txt",
+    content.replace(/^\uFEFF/, "").replace(/\r/g, "").split("\n").map((text) => ({ start: null, end: null, text }))
+  );
+}
+function objectValue(obj, keys) {
+  for (const key of keys) {
+    if (key in obj) return obj[key];
+  }
+  return void 0;
+}
+function lineFromUnknown(value) {
+  if (typeof value === "string") {
+    const text2 = cleanText(value);
+    return text2 ? { start: null, end: null, text: text2 } : null;
+  }
+  if (!value || typeof value !== "object" || Array.isArray(value)) return null;
+  const obj = value;
+  const text = cleanText(objectValue(obj, ["text", "phrase", "content", "line", "lyrics", "lyric"]));
+  if (!text) return null;
+  const start = toSeconds(objectValue(obj, ["start", "start_time", "startTime", "begin", "from", "timestamp"]));
+  let end = toSeconds(objectValue(obj, ["end", "end_time", "endTime", "stop", "to"]));
+  if (end === null && start !== null) {
+    const duration = toSeconds(objectValue(obj, ["duration", "length"]));
+    if (duration !== null) end = start + duration;
+  }
+  const type = cleanText(objectValue(obj, ["type", "kind", "segment_type", "segmentType"]));
+  return type ? { start, end, text, type } : { start, end, text };
+}
+function findJsonLineArray(value) {
+  if (Array.isArray(value)) return Array.from(value);
+  if (!value || typeof value !== "object") return null;
+  const obj = value;
+  for (const key of ["phrases", "phrase_segments", "phraseSegments", "segments", "chunks", "lines", "items", "transcript"]) {
+    const candidate = obj[key];
+    if (Array.isArray(candidate)) return Array.from(candidate);
+  }
+  return null;
+}
+function looksLikePhraseJson(value) {
+  if (!value || typeof value !== "object" || Array.isArray(value)) return false;
+  const obj = value;
+  if (["phrases", "phrase_segments", "phraseSegments", "chunks"].some((key) => Array.isArray(obj[key]))) {
+    return true;
+  }
+  const engineTarget = cleanText(obj.engine_version_target).toLowerCase();
+  if (engineTarget.includes("music transcription")) return true;
+  const segments = obj.segments;
+  if (!Array.isArray(segments)) return false;
+  return segments.some((segment) => {
+    if (!segment || typeof segment !== "object" || Array.isArray(segment)) return false;
+    const segmentObj = segment;
+    const type = cleanText(segmentObj.type).toLowerCase();
+    return type === "lyrics" || type === "music";
+  });
+}
+function parseJson(content, requestedFormat) {
+  const parsed = JSON.parse(content);
+  const format = requestedFormat ?? (looksLikePhraseJson(parsed) ? "phrase-json" : "json");
+  const values = findJsonLineArray(parsed);
+  if (!values) {
+    const single = lineFromUnknown(parsed);
+    return buildTranscript(format, single ? [single] : []);
+  }
+  return buildTranscript(
+    format,
+    values.map(lineFromUnknown).filter((line) => line !== null)
+  );
+}
+function detectTranscriptFormat(content, filename) {
+  const lowerName = filename?.toLowerCase() ?? "";
+  if (lowerName.endsWith(".phrase-timing.json") || lowerName.endsWith(".phrase.json") || lowerName.endsWith(".phrases.json")) {
+    return "phrase-json";
+  }
+  if (lowerName.endsWith(".srt")) return "srt";
+  if (lowerName.endsWith(".vtt")) return "vtt";
+  if (lowerName.endsWith(".lrc")) return "lrc";
+  if (lowerName.endsWith(".txt")) return "txt";
+  if (lowerName.endsWith(".json")) {
+    try {
+      const parsed = JSON.parse(content);
+      return looksLikePhraseJson(parsed) ? "phrase-json" : "json";
+    } catch {
+      return "json";
+    }
+  }
+  const trimmed = content.trim();
+  if (/^WEBVTT\b/.test(trimmed)) return "vtt";
+  if (/\d{1,2}:\d{2}:\d{2}[,.]\d{3}\s+-->\s+\d{1,2}:\d{2}:\d{2}[,.]\d{3}/.test(trimmed)) return "srt";
+  if (/^\s*\[\d{1,3}:\d{2}(?:\.\d{1,3})?\]/m.test(trimmed)) return "lrc";
+  if (trimmed.startsWith("{") && trimmed.endsWith("}") || trimmed.startsWith("[") && trimmed.endsWith("]")) {
+    try {
+      const parsed = JSON.parse(trimmed);
+      return looksLikePhraseJson(parsed) ? "phrase-json" : "json";
+    } catch {
+    }
+  }
+  return "txt";
+}
+function parseTranscript(content, options = {}) {
+  const format = options.format ?? detectTranscriptFormat(content, options.filename);
+  switch (format) {
+    case "srt":
+      return parseTimedBlocks(content, "srt");
+    case "vtt":
+      return parseTimedBlocks(content, "vtt");
+    case "lrc":
+      return parseLrc(content);
+    case "txt":
+      return parseTxt(content);
+    case "json":
+      return parseJson(content, "json");
+    case "phrase-json":
+      return parseJson(content, "phrase-json");
+  }
+}
+
+exports.detectTranscriptFormat = detectTranscriptFormat;
+exports.parseTranscript = parseTranscript;
+//# sourceMappingURL=index.cjs.map
+//# sourceMappingURL=index.cjs.map

package/dist/index.cjs.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/index.ts"],"names":["text"],"mappings":";;;AA0BA,SAAS,UAAU,KAAA,EAAwB;AACzC,EAAA,OAAO,OAAO,UAAU,QAAA,GAAW,KAAA,CAAM,QAAQ,MAAA,EAAQ,GAAG,CAAA,CAAE,IAAA,EAAK,GAAI,EAAA;AACzE;AAEA,SAAS,UAAU,KAAA,EAA+B;AAChD,EAAA,IAAI,OAAO,KAAA,KAAU,QAAA,IAAY,OAAO,QAAA,CAAS,KAAK,GAAG,OAAO,KAAA;AAChE,EAAA,IAAI,OAAO,KAAA,KAAU,QAAA,EAAU,OAAO,IAAA;AAEtC,EAAA,MAAM,aAAa,KAAA,CAAM,IAAA,EAAK,CAAE,OAAA,CAAQ,KAAK,GAAG,CAAA;AAChD,EAAA,IAAI,CAAC,YAAY,OAAO,IAAA;AAExB,EAAA,IAAI,iBAAA,CAAkB,IAAA,CAAK,UAAU,CAAA,EAAG;AACtC,IAAA,MAAM,MAAA,GAAS,OAAO,UAAU,CAAA;AAChC,IAAA,OAAO,MAAA,CAAO,QAAA,CAAS,MAAM,CAAA,GAAI,MAAA,GAAS,IAAA;AAAA,EAC5C;AAEA,EAAA,MAAM,QAAQ,UAAA,CAAW,KAAA,CAAM,GAAG,CAAA,CAAE,IAAI,MAAM,CAAA;AAC9C,EAAA,IAAI,KAAA,CAAM,IAAA,CAAK,CAAC,IAAA,KAAS,CAAC,OAAO,QAAA,CAAS,IAAI,CAAC,CAAA,EAAG,OAAO,IAAA;AAEzD,EAAA,IAAI,KAAA,CAAM,MAAA,KAAW,CAAA,EAAG,OAAO,KAAA,CAAM,CAAC,CAAA,GAAK,IAAA,GAAO,KAAA,CAAM,CAAC,CAAA,GAAK,EAAA,GAAK,MAAM,CAAC,CAAA;AAC1E,EAAA,IAAI,KAAA,CAAM,WAAW,CAAA,EAAG,OAAO,MAAM,CAAC,CAAA,GAAK,EAAA,GAAK,KAAA,CAAM,CAAC,CAAA;AAEvD,EAAA,OAAO,IAAA;AACT;AAEA,SAAS,eAAA,CAAgB,QAA0B,KAAA,EAAqC;AACtF,EAAA,MAAM,QAAA,GAAW,MACd,GAAA,CAAI,CAAC,UAAU,EAAE,GAAG,MAAM,IAAA,EAAM,SAAA,CAAU,KAAK,IAAI,CAAA,GAAI,CAAA,CACvD,MAAA,CAAO,CAAC,IAAA,KAAS,IAAA,CAAK,IAAA,CAAK,MAAA,GAAS,CAAC,CAAA;AAExC,EAAA,OAAO;AAAA,IACL,MAAA;AAAA,IACA,KAAA,EAAO,QAAA;AAAA,IACP,IAAA,EAAM,SAAS,GAAA,CAAI,CAAC,SAAS,IAAA,CAAK,IAAI,CAAA,CAAE,IAAA,CAAK,IAAI;AAAA,GACnD;AACF;AAEA,SAAS,gBAAA,CAAiB,SAAiB,MAAA,EAAmC;AAC5E,EAAA,MAAM,UAAA,GAAa,QAAQ,OAAA,CAAQ,SAAA,EAAW,EAAE,CAAA,CAAE,OAAA,CAAQ,OAAO,EAAE,CAAA;AACnE,EAAA,MAAM,OAAO,MAAA,KAAW,KAAA,GAAQ,WAAW,OAAA,CAAQ,kBAAA,EAAoB,EAAE,CAAA,GAAI,UAAA;AAC7E,EAAA,MAAM,MAAA,GAAS,IAAA,CAAK,KAAA,CAAM,QAAQ,CAAA;AAClC,EAAA,MAAM,QAA0B,EAAC;AAEjC,EAAA,KAAA,MAAW,SAAS,MAAA,EAAQ;AAC1B,IAAA,MAAM,OAAO,KAAA,CAAM,KAAA,CAAM,IAAI,CAAA,CAAE,OAAO,OAAO,CAAA;AAC7C,IAAA,IAAI,IAAA,CAAK,SAAS,CAAA,EAAG;AAErB,IAAA,MAAM,WAAA,GAAc,KAAK,SAAA,CAAU,CAAC,QAAQ,GAAA,CAAI,QAAA,CAAS,KAAK,CAAC,CAAA;AAC/D,IAAA,IAAI,cAAc,CAAA,EAAG;AAErB,IAAA,MAAM,MAAA,GAAS,IAAA,CAAK,WAAW,CAAA,CAAG,MAAM,KAAK,CAAA;AAC7C,IAAA,MAAM,KAAA,GAAQ,SAAA,CAAU,MAAA,CAAO,CAAC,CAAC,CAAA;AACjC,IAAA,MAAM,GAAA,GAAM,SAAA,CAAU,MAAA,CAAO,CAAC,CAAA,EAAG,IAAA,EAAK,CAAE,KAAA,CAAM,KAAK,CAAA,CAAE,CAAC,CAAC,CAAA;AACvD,IAAA,MAAM,OAAO,IAAA,CAAK,KAAA,CAAM,cAAc,CAAC,CAAA,CAAE,KAAK,GAAG,CAAA;AAEjD,IAAA,IAAI,UAAU,IAAA,IAAQ,GAAA,KAAQ,IAAA,IAAQ,SAAA,CAAU,IAAI,CAAA,EAAG;AACrD,MAAA,KAAA,CAAM,IAAA,CAAK,EAAE,KAAA,EAAO,GAAA,EAAK,MAAM,CAAA;AAAA,IACjC;AAAA,EACF;AAEA,EAAA,OAAO,eAAA,CAAgB,QAAQ,KAAK,CAAA;AACtC;AAEA,SAAS,SAAS,OAAA,EAA6B;AAC7C,EAAA,MAAM,QAA0B,EAAC;AAEjC,EAAA,KAAA,MAAW,OAAA,IAAW,QAAQ,OAAA,CAAQ,KAAA,EAAO,EAAE,CAAA,CAAE,KAAA,CAAM,IAAI,CAAA,EAAG;AAC5D,IAAA,MAAM,UAAU,CAAC,GAAG,OAAA,CAAQ,QAAA,CAAS,sCAAsC,CAAC,CAAA;AAC5E,IAAA,MAAM,OAAO,OAAA,CAAQ,OAAA,CAAQ,aAAA,EAAe,EAAE,EAAE,IAAA,EAAK;AACrD,IAAA,IAAI,CAAC,IAAA,IAAQ,OAAA,CAAQ,MAAA,KAAW,CAAA,EAAG;AAEnC,IAAA,KAAA,MAAW,SAAS,OAAA,EAAS;AAC3B,MAAA,MAAM,OAAA,GAAU,MAAA,CAAO,KAAA,CAAM,CAAC,CAAC,CAAA;AAC/B,MAAA,MAAM,OAAA,GAAU,MAAA,CAAO,KAAA,CAAM,CAAC,CAAC,CAAA;AAC/B,MAAA,KAAA,CAAM,IAAA,CAAK,EAAE,KAAA,EAAO,OAAA,GAAU,KAAK,OAAA,EAAS,GAAA,EAAK,IAAA,EAAM,IAAA,EAAM,CAAA;AAAA,IAC/D;AAAA,EACF;AAEA,EAAA,KAAA,CAAM,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAA,CAAO,EAAE,KAAA,IAAS,CAAA,KAAM,CAAA,CAAE,KAAA,IAAS,CAAA,CAAE,CAAA;AACpD,EAAA,KAAA,IAAS,IAAI,CAAA,EAAG,CAAA,GAAI,MAAM,MAAA,GAAS,CAAA,EAAG,KAAK,CAAA,EAAG;AAC5C,IAAA,KAAA,CAAM,CAAC,CAAA,CAAG,GAAA,GAAM,KAAA,CAAM,CAAA,GAAI,CAAC,CAAA,CAAG,KAAA;AAAA,EAChC;AAEA,EAAA,OAAO,eAAA,CAAgB,OAAO,KAAK,CAAA;AACrC;AAEA,SAAS,SAAS,OAAA,EAA6B;AAC7C,EAAA,OAAO,eAAA;AAAA,IACL,KAAA;AAAA,IACA,OAAA,CACG,QAAQ,SAAA,EAAW,EAAE,EACrB,OAAA,CAAQ,KAAA,EAAO,EAAE,CAAA,CACjB,KAAA,CAAM,IAAI,CAAA,CACV,GAAA,CAAI,CAAC,IAAA,MAAU,EAAE,OAAO,IAAA,EAAM,GAAA,EAAK,IAAA,EAAM,IAAA,EAAK,CAAE;AAAA,GACrD;AACF;AAEA,SAAS,WAAA,CAAY,KAA8B,IAAA,EAAyB;AAC1E,EAAA,KAAA,MAAW,OAAO,IAAA,EAAM;AACtB,IAAA,IAAI,GAAA,IAAO,GAAA,EAAK,OAAO,GAAA,CAAI,GAAG,CAAA;AAAA,EAChC;AACA,EAAA,OAAO,MAAA;AACT;AAEA,SAAS,gBAAgB,KAAA,EAAuC;AAC9D,EAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,IAAA,MAAMA,KAAAA,GAAO,UAAU,KAAK,CAAA;AAC5B,IAAA,OAAOA,KAAAA,GAAO,EAAE,KAAA,EAAO,IAAA,EAAM,KAAK,IAAA,EAAM,IAAA,EAAAA,OAAK,GAAI,IAAA;AAAA,EACnD;AAEA,EAAA,IAAI,CAAC,SAAS,OAAO,KAAA,KAAU,YAAY,KAAA,CAAM,OAAA,CAAQ,KAAK,CAAA,EAAG,OAAO,IAAA;AACxE,EAAA,MAAM,GAAA,GAAM,KAAA;AACZ,EAAA,MAAM,IAAA,GAAO,SAAA,CAAU,WAAA,CAAY,GAAA,EAAK,CAAC,MAAA,EAAQ,QAAA,EAAU,SAAA,EAAW,MAAA,EAAQ,QAAA,EAAU,OAAO,CAAC,CAAC,CAAA;AACjG,EAAA,IAAI,CAAC,MAAM,OAAO,IAAA;AAElB,EAAA,MAAM,KAAA,GAAQ,SAAA,CAAU,WAAA,CAAY,GAAA,EAAK,CAAC,OAAA,EAAS,YAAA,EAAc,WAAA,EAAa,OAAA,EAAS,MAAA,EAAQ,WAAW,CAAC,CAAC,CAAA;AAC5G,EAAA,IAAI,GAAA,GAAM,SAAA,CAAU,WAAA,CAAY,GAAA,EAAK,CAAC,KAAA,EAAO,UAAA,EAAY,SAAA,EAAW,MAAA,EAAQ,IAAI,CAAC,CAAC,CAAA;AAElF,EAAA,IAAI,GAAA,KAAQ,IAAA,IAAQ,KAAA,KAAU,IAAA,EAAM;AAClC,IAAA,MAAM,QAAA,GAAW,UAAU,WAAA,CAAY,GAAA,EAAK,CAAC,UAAA,EAAY,QAAQ,CAAC,CAAC,CAAA;AACnE,IAAA,IAAI,QAAA,KAAa,IAAA,EAAM,GAAA,GAAM,KAAA,GAAQ,QAAA;AAAA,EACvC;AAEA,EAAA,MAAM,IAAA,GAAO,SAAA,CAAU,WAAA,CAAY,GAAA,EAAK,CAAC,QAAQ,MAAA,EAAQ,cAAA,EAAgB,aAAa,CAAC,CAAC,CAAA;AACxF,EAAA,OAAO,IAAA,GAAO,EAAE,KAAA,EAAO,GAAA,EAAK,IAAA,EAAM,MAAK,GAAI,EAAE,KAAA,EAAO,GAAA,EAAK,IAAA,EAAK;AAChE;AAEA,SAAS,kBAAkB,KAAA,EAAkC;AAC3D,EAAA,IAAI,MAAM,OAAA,CAAQ,KAAK,GAAG,OAAO,KAAA,CAAM,KAAK,KAAkB,CAAA;AAC9D,EAAA,IAAI,CAAC,KAAA,IAAS,OAAO,KAAA,KAAU,UAAU,OAAO,IAAA;AAEhD,EAAA,MAAM,GAAA,GAAM,KAAA;AACZ,EAAA,KAAA,MAAW,GAAA,IAAO,CAAC,SAAA,EAAW,iBAAA,EAAmB,gBAAA,EAAkB,YAAY,QAAA,EAAU,OAAA,EAAS,OAAA,EAAS,YAAY,CAAA,EAAG;AACxH,IAAA,MAAM,SAAA,GAAY,IAAI,GAAG,CAAA;AACzB,IAAA,IAAI,MAAM,OAAA,CAAQ,SAAS,GAAG,OAAO,KAAA,CAAM,KAAK,SAAsB,CAAA;AAAA,EACxE;AAEA,EAAA,OAAO,IAAA;AACT;AAEA,SAAS,oBAAoB,KAAA,EAAyB;AACpD,EAAA,IAAI,CAAC,SAAS,OAAO,KAAA,KAAU,YAAY,KAAA,CAAM,OAAA,CAAQ,KAAK,CAAA,EAAG,OAAO,KAAA;AACxE,EAAA,MAAM,GAAA,GAAM,KAAA;AAEZ,EAAA,IAAI,CAAC,SAAA,EAAW,iBAAA,EAAmB,gBAAA,EAAkB,QAAQ,CAAA,CAAE,IAAA,CAAK,CAAC,GAAA,KAAQ,MAAM,OAAA,CAAQ,GAAA,CAAI,GAAG,CAAC,CAAC,CAAA,EAAG;AACrG,IAAA,OAAO,IAAA;AAAA,EACT;AAEA,EAAA,MAAM,YAAA,GAAe,SAAA,CAAU,GAAA,CAAI,qBAAqB,EAAE,WAAA,EAAY;AACtE,EAAA,IAAI,YAAA,CAAa,QAAA,CAAS,qBAAqB,CAAA,EAAG,OAAO,IAAA;AAEzD,EAAA,MAAM,WAAW,GAAA,CAAI,QAAA;AACrB,EAAA,IAAI,CAAC,KAAA,CAAM,OAAA,CAAQ,QAAQ,GAAG,OAAO,KAAA;AAErC,EAAA,OAAO,QAAA,CAAS,IAAA,CAAK,CAAC,OAAA,KAAqB;AACzC,IAAA,IAAI,CAAC,WAAW,OAAO,OAAA,KAAY,YAAY,KAAA,CAAM,OAAA,CAAQ,OAAO,CAAA,EAAG,OAAO,KAAA;AAC9E,IAAA,MAAM,UAAA,GAAa,OAAA;AACnB,IAAA,MAAM,IAAA,GAAO,SAAA,CAAU,UAAA,CAAW,IAAI,EAAE,WAAA,EAAY;AACpD,IAAA,OAAO,IAAA,KAAS,YAAY,IAAA,KAAS,OAAA;AAAA,EACvC,CAAC,CAAA;AACH;AAEA,SAAS,SAAA,CAAU,SAAiB,eAAA,EAAsD;AACxF,EAAA,MAAM,MAAA,GAAkB,IAAA,CAAK,KAAA,CAAM,OAAO,CAAA;AAC1C,EAAA,MAAM,MAAA,GAAiC,eAAA,KAAoB,mBAAA,CAAoB,MAAM,IAAI,aAAA,GAAgB,MAAA,CAAA;AACzG,EAAA,MAAM,MAAA,GAAS,kBAAkB,MAAM,CAAA;AAEvC,EAAA,IAAI,CAAC,MAAA,EAAQ;AACX,IAAA,MAAM,MAAA,GAAS,gBAAgB,MAAM,CAAA;AACrC,IAAA,OAAO,gBAAgB,MAAA,EAAQ,MAAA,GAAS,CAAC,MAAM,CAAA,GAAI,EAAE,CAAA;AAAA,EACvD;AAEA,EAAA,OAAO,eAAA;AAAA,IACL,MAAA;AAAA,IACA,MAAA,CAAO,IAAI,eAAe,CAAA,CAAE,OAAO,CAAC,IAAA,KAAiC,SAAS,IAAI;AAAA,GACpF;AACF;AAEO,SAAS,sBAAA,CAAuB,SAAiB,QAAA,EAAqC;AAC3F,EAAA,MAAM,SAAA,GAAY,QAAA,EAAU,WAAA,EAAY,IAAK,EAAA;AAE7C,EAAA,IACE,SAAA,CAAU,QAAA,CAAS,qBAAqB,CAAA,IACxC,SAAA,CAAU,QAAA,CAAS,cAAc,CAAA,IACjC,SAAA,CAAU,QAAA,CAAS,eAAe,CAAA,EAClC;AACA,IAAA,OAAO,aAAA;AAAA,EACT;AACA,EAAA,IAAI,SAAA,CAAU,QAAA,CAAS,MAAM,CAAA,EAAG,OAAO,KAAA;AACvC,EAAA,IAAI,SAAA,CAAU,QAAA,CAAS,MAAM,CAAA,EAAG,OAAO,KAAA;AACvC,EAAA,IAAI,SAAA,CAAU,QAAA,CAAS,MAAM,CAAA,EAAG,OAAO,KAAA;AACvC,EAAA,IAAI,SAAA,CAAU,QAAA,CAAS,MAAM,CAAA,EAAG,OAAO,KAAA;AACvC,EAAA,IAAI,SAAA,CAAU,QAAA,CAAS,OAAO,CAAA,EAAG;AAC/B,IAAA,IAAI;AACF,MAAA,MAAM,MAAA,GAAkB,IAAA,CAAK,KAAA,CAAM,OAAO,CAAA;AAC1C,MAAA,OAAO,mBAAA,CAAoB,MAAM,CAAA,GAAI,aAAA,GAAgB,MAAA;AAAA,IACvD,CAAA,CAAA,MAAQ;AACN,MAAA,OAAO,MAAA;AAAA,IACT;AAAA,EACF;AAEA,EAAA,MAAM,OAAA,GAAU,QAAQ,IAAA,EAAK;AAC7B,EAAA,IAAI,WAAA,CAAY,IAAA,CAAK,OAAO,CAAA,EAAG,OAAO,KAAA;AACtC,EAAA,IAAI,mEAAA,CAAoE,IAAA,CAAK,OAAO,CAAA,EAAG,OAAO,KAAA;AAC9F,EAAA,IAAI,sCAAA,CAAuC,IAAA,CAAK,OAAO,CAAA,EAAG,OAAO,KAAA;AAEjE,EAAA,IAAK,OAAA,CAAQ,UAAA,CAAW,GAAG,CAAA,IAAK,QAAQ,QAAA,CAAS,GAAG,CAAA,IAAO,OAAA,CAAQ,WAAW,GAAG,CAAA,IAAK,OAAA,CAAQ,QAAA,CAAS,GAAG,CAAA,EAAI;AAC5G,IAAA,IAAI;AACF,MAAA,MAAM,MAAA,GAAkB,IAAA,CAAK,KAAA,CAAM,OAAO,CAAA;AAC1C,MAAA,OAAO,mBAAA,CAAoB,MAAM,CAAA,GAAI,aAAA,GAAgB,MAAA;AAAA,IACvD,CAAA,CAAA,MAAQ;AAAA,IAER;AAAA,EACF;AAEA,EAAA,OAAO,KAAA;AACT;AAEO,SAAS,eAAA,CAAgB,OAAA,EAAiB,OAAA,GAAkC,EAAC,EAAe;AACjG,EAAA,MAAM,SAAS,OAAA,CAAQ,MAAA,IAAU,sBAAA,CAAuB,OAAA,EAAS,QAAQ,QAAQ,CAAA;AAEjF,EAAA,QAAQ,MAAA;AAAQ,IACd,KAAK,KAAA;AACH,MAAA,OAAO,gBAAA,CAAiB,SAAS,KAAK,CAAA;AAAA,IACxC,KAAK,KAAA;AACH,MAAA,OAAO,gBAAA,CAAiB,SAAS,KAAK,CAAA;AAAA,IACxC,KAAK,KAAA;AACH,MAAA,OAAO,SAAS,OAAO,CAAA;AAAA,IACzB,KAAK,KAAA;AACH,MAAA,OAAO,SAAS,OAAO,CAAA;AAAA,IACzB,KAAK,MAAA;AACH,MAAA,OAAO,SAAA,CAAU,SAAS,MAAM,CAAA;AAAA,IAClC,KAAK,aAAA;AACH,MAAA,OAAO,SAAA,CAAU,SAAS,aAAa,CAAA;AAAA;AAE7C","file":"index.cjs","sourcesContent":["export type TranscriptFormat =\n  | 'srt'\n  | 'vtt'\n  | 'lrc'\n  | 'txt'\n  | 'json'\n  | 'phrase-json';\n\nexport interface TranscriptLine {\n  start: number | null;\n  end: number | null;\n  text: string;\n  type?: string;\n}\n\nexport interface Transcript {\n  format: TranscriptFormat;\n  lines: TranscriptLine[];\n  text: string;\n}\n\nexport interface ParseTranscriptOptions {\n  format?: TranscriptFormat;\n  filename?: string;\n}\n\nfunction cleanText(value: unknown): string {\n  return typeof value === 'string' ? value.replace(/\\s+/g, ' ').trim() : '';\n}\n\nfunction toSeconds(value: unknown): number | null {\n  if (typeof value === 'number' && Number.isFinite(value)) return value;\n  if (typeof value !== 'string') return null;\n\n  const normalized = value.trim().replace(',', '.');\n  if (!normalized) return null;\n\n  if (/^\\d+(?:\\.\\d+)?$/.test(normalized)) {\n    const parsed = Number(normalized);\n    return Number.isFinite(parsed) ? parsed : null;\n  }\n\n  const parts = normalized.split(':').map(Number);\n  if (parts.some((part) => !Number.isFinite(part))) return null;\n\n  if (parts.length === 3) return parts[0]! * 3600 + parts[1]! * 60 + parts[2]!;\n  if (parts.length === 2) return parts[0]! * 60 + parts[1]!;\n\n  return null;\n}\n\nfunction buildTranscript(format: TranscriptFormat, lines: TranscriptLine[]): Transcript {\n  const filtered = lines\n    .map((line) => ({ ...line, text: cleanText(line.text) }))\n    .filter((line) => line.text.length > 0);\n\n  return {\n    format,\n    lines: filtered,\n    text: filtered.map((line) => line.text).join('\\n')\n  };\n}\n\nfunction parseTimedBlocks(content: string, format: 'srt' | 'vtt'): Transcript {\n  const normalized = content.replace(/^\\uFEFF/, '').replace(/\\r/g, '');\n  const body = format === 'vtt' ? normalized.replace(/^WEBVTT[^\\n]*\\n+/, '') : normalized;\n  const blocks = body.split(/\\n{2,}/);\n  const lines: TranscriptLine[] = [];\n\n  for (const block of blocks) {\n    const rows = block.split('\\n').filter(Boolean);\n    if (rows.length < 2) continue;\n\n    const timingIndex = rows.findIndex((row) => row.includes('-->'));\n    if (timingIndex < 0) continue;\n\n    const timing = rows[timingIndex]!.split('-->');\n    const start = toSeconds(timing[0]);\n    const end = toSeconds(timing[1]?.trim().split(/\\s+/)[0]);\n    const text = rows.slice(timingIndex + 1).join(' ');\n\n    if (start !== null || end !== null || cleanText(text)) {\n      lines.push({ start, end, text });\n    }\n  }\n\n  return buildTranscript(format, lines);\n}\n\nfunction parseLrc(content: string): Transcript {\n  const lines: TranscriptLine[] = [];\n\n  for (const rawLine of content.replace(/\\r/g, '').split('\\n')) {\n    const matches = [...rawLine.matchAll(/\\[(\\d{1,3}):(\\d{2}(?:\\.\\d{1,3})?)\\]/g)];\n    const text = rawLine.replace(/\\[[^\\]]+\\]/g, '').trim();\n    if (!text || matches.length === 0) continue;\n\n    for (const match of matches) {\n      const minutes = Number(match[1]);\n      const seconds = Number(match[2]);\n      lines.push({ start: minutes * 60 + seconds, end: null, text });\n    }\n  }\n\n  lines.sort((a, b) => (a.start ?? 0) - (b.start ?? 0));\n  for (let i = 0; i < lines.length - 1; i += 1) {\n    lines[i]!.end = lines[i + 1]!.start;\n  }\n\n  return buildTranscript('lrc', lines);\n}\n\nfunction parseTxt(content: string): Transcript {\n  return buildTranscript(\n    'txt',\n    content\n      .replace(/^\\uFEFF/, '')\n      .replace(/\\r/g, '')\n      .split('\\n')\n      .map((text) => ({ start: null, end: null, text }))\n  );\n}\n\nfunction objectValue(obj: Record<string, unknown>, keys: string[]): unknown {\n  for (const key of keys) {\n    if (key in obj) return obj[key];\n  }\n  return undefined;\n}\n\nfunction lineFromUnknown(value: unknown): TranscriptLine | null {\n  if (typeof value === 'string') {\n    const text = cleanText(value);\n    return text ? { start: null, end: null, text } : null;\n  }\n\n  if (!value || typeof value !== 'object' || Array.isArray(value)) return null;\n  const obj = value as Record<string, unknown>;\n  const text = cleanText(objectValue(obj, ['text', 'phrase', 'content', 'line', 'lyrics', 'lyric']));\n  if (!text) return null;\n\n  const start = toSeconds(objectValue(obj, ['start', 'start_time', 'startTime', 'begin', 'from', 'timestamp']));\n  let end = toSeconds(objectValue(obj, ['end', 'end_time', 'endTime', 'stop', 'to']));\n\n  if (end === null && start !== null) {\n    const duration = toSeconds(objectValue(obj, ['duration', 'length']));\n    if (duration !== null) end = start + duration;\n  }\n\n  const type = cleanText(objectValue(obj, ['type', 'kind', 'segment_type', 'segmentType']));\n  return type ? { start, end, text, type } : { start, end, text };\n}\n\nfunction findJsonLineArray(value: unknown): unknown[] | null {\n  if (Array.isArray(value)) return Array.from(value as unknown[]);\n  if (!value || typeof value !== 'object') return null;\n\n  const obj = value as Record<string, unknown>;\n  for (const key of ['phrases', 'phrase_segments', 'phraseSegments', 'segments', 'chunks', 'lines', 'items', 'transcript']) {\n    const candidate = obj[key];\n    if (Array.isArray(candidate)) return Array.from(candidate as unknown[]);\n  }\n\n  return null;\n}\n\nfunction looksLikePhraseJson(value: unknown): boolean {\n  if (!value || typeof value !== 'object' || Array.isArray(value)) return false;\n  const obj = value as Record<string, unknown>;\n\n  if (['phrases', 'phrase_segments', 'phraseSegments', 'chunks'].some((key) => Array.isArray(obj[key]))) {\n    return true;\n  }\n\n  const engineTarget = cleanText(obj.engine_version_target).toLowerCase();\n  if (engineTarget.includes('music transcription')) return true;\n\n  const segments = obj.segments;\n  if (!Array.isArray(segments)) return false;\n\n  return segments.some((segment: unknown) => {\n    if (!segment || typeof segment !== 'object' || Array.isArray(segment)) return false;\n    const segmentObj = segment as Record<string, unknown>;\n    const type = cleanText(segmentObj.type).toLowerCase();\n    return type === 'lyrics' || type === 'music';\n  });\n}\n\nfunction parseJson(content: string, requestedFormat?: 'json' | 'phrase-json'): Transcript {\n  const parsed: unknown = JSON.parse(content);\n  const format: 'json' | 'phrase-json' = requestedFormat ?? (looksLikePhraseJson(parsed) ? 'phrase-json' : 'json');\n  const values = findJsonLineArray(parsed);\n\n  if (!values) {\n    const single = lineFromUnknown(parsed);\n    return buildTranscript(format, single ? [single] : []);\n  }\n\n  return buildTranscript(\n    format,\n    values.map(lineFromUnknown).filter((line): line is TranscriptLine => line !== null)\n  );\n}\n\nexport function detectTranscriptFormat(content: string, filename?: string): TranscriptFormat {\n  const lowerName = filename?.toLowerCase() ?? '';\n\n  if (\n    lowerName.endsWith('.phrase-timing.json') ||\n    lowerName.endsWith('.phrase.json') ||\n    lowerName.endsWith('.phrases.json')\n  ) {\n    return 'phrase-json';\n  }\n  if (lowerName.endsWith('.srt')) return 'srt';\n  if (lowerName.endsWith('.vtt')) return 'vtt';\n  if (lowerName.endsWith('.lrc')) return 'lrc';\n  if (lowerName.endsWith('.txt')) return 'txt';\n  if (lowerName.endsWith('.json')) {\n    try {\n      const parsed: unknown = JSON.parse(content);\n      return looksLikePhraseJson(parsed) ? 'phrase-json' : 'json';\n    } catch {\n      return 'json';\n    }\n  }\n\n  const trimmed = content.trim();\n  if (/^WEBVTT\\b/.test(trimmed)) return 'vtt';\n  if (/\\d{1,2}:\\d{2}:\\d{2}[,.]\\d{3}\\s+-->\\s+\\d{1,2}:\\d{2}:\\d{2}[,.]\\d{3}/.test(trimmed)) return 'srt';\n  if (/^\\s*\\[\\d{1,3}:\\d{2}(?:\\.\\d{1,3})?\\]/m.test(trimmed)) return 'lrc';\n\n  if ((trimmed.startsWith('{') && trimmed.endsWith('}')) || (trimmed.startsWith('[') && trimmed.endsWith(']'))) {\n    try {\n      const parsed: unknown = JSON.parse(trimmed);\n      return looksLikePhraseJson(parsed) ? 'phrase-json' : 'json';\n    } catch {\n      // fall through to txt\n    }\n  }\n\n  return 'txt';\n}\n\nexport function parseTranscript(content: string, options: ParseTranscriptOptions = {}): Transcript {\n  const format = options.format ?? detectTranscriptFormat(content, options.filename);\n\n  switch (format) {\n    case 'srt':\n      return parseTimedBlocks(content, 'srt');\n    case 'vtt':\n      return parseTimedBlocks(content, 'vtt');\n    case 'lrc':\n      return parseLrc(content);\n    case 'txt':\n      return parseTxt(content);\n    case 'json':\n      return parseJson(content, 'json');\n    case 'phrase-json':\n      return parseJson(content, 'phrase-json');\n  }\n}\n"]}

package/dist/index.d.cts

@@ -0,0 +1,20 @@
+type TranscriptFormat = 'srt' | 'vtt' | 'lrc' | 'txt' | 'json' | 'phrase-json';
+interface TranscriptLine {
+    start: number | null;
+    end: number | null;
+    text: string;
+    type?: string;
+}
+interface Transcript {
+    format: TranscriptFormat;
+    lines: TranscriptLine[];
+    text: string;
+}
+interface ParseTranscriptOptions {
+    format?: TranscriptFormat;
+    filename?: string;
+}
+declare function detectTranscriptFormat(content: string, filename?: string): TranscriptFormat;
+declare function parseTranscript(content: string, options?: ParseTranscriptOptions): Transcript;
+
+export { type ParseTranscriptOptions, type Transcript, type TranscriptFormat, type TranscriptLine, detectTranscriptFormat, parseTranscript };

package/dist/index.d.ts

@@ -0,0 +1,20 @@
+type TranscriptFormat = 'srt' | 'vtt' | 'lrc' | 'txt' | 'json' | 'phrase-json';
+interface TranscriptLine {
+    start: number | null;
+    end: number | null;
+    text: string;
+    type?: string;
+}
+interface Transcript {
+    format: TranscriptFormat;
+    lines: TranscriptLine[];
+    text: string;
+}
+interface ParseTranscriptOptions {
+    format?: TranscriptFormat;
+    filename?: string;
+}
+declare function detectTranscriptFormat(content: string, filename?: string): TranscriptFormat;
+declare function parseTranscript(content: string, options?: ParseTranscriptOptions): Transcript;
+
+export { type ParseTranscriptOptions, type Transcript, type TranscriptFormat, type TranscriptLine, detectTranscriptFormat, parseTranscript };

package/dist/index.js

@@ -0,0 +1,186 @@
+// src/index.ts
+function cleanText(value) {
+  return typeof value === "string" ? value.replace(/\s+/g, " ").trim() : "";
+}
+function toSeconds(value) {
+  if (typeof value === "number" && Number.isFinite(value)) return value;
+  if (typeof value !== "string") return null;
+  const normalized = value.trim().replace(",", ".");
+  if (!normalized) return null;
+  if (/^\d+(?:\.\d+)?$/.test(normalized)) {
+    const parsed = Number(normalized);
+    return Number.isFinite(parsed) ? parsed : null;
+  }
+  const parts = normalized.split(":").map(Number);
+  if (parts.some((part) => !Number.isFinite(part))) return null;
+  if (parts.length === 3) return parts[0] * 3600 + parts[1] * 60 + parts[2];
+  if (parts.length === 2) return parts[0] * 60 + parts[1];
+  return null;
+}
+function buildTranscript(format, lines) {
+  const filtered = lines.map((line) => ({ ...line, text: cleanText(line.text) })).filter((line) => line.text.length > 0);
+  return {
+    format,
+    lines: filtered,
+    text: filtered.map((line) => line.text).join("\n")
+  };
+}
+function parseTimedBlocks(content, format) {
+  const normalized = content.replace(/^\uFEFF/, "").replace(/\r/g, "");
+  const body = format === "vtt" ? normalized.replace(/^WEBVTT[^\n]*\n+/, "") : normalized;
+  const blocks = body.split(/\n{2,}/);
+  const lines = [];
+  for (const block of blocks) {
+    const rows = block.split("\n").filter(Boolean);
+    if (rows.length < 2) continue;
+    const timingIndex = rows.findIndex((row) => row.includes("-->"));
+    if (timingIndex < 0) continue;
+    const timing = rows[timingIndex].split("-->");
+    const start = toSeconds(timing[0]);
+    const end = toSeconds(timing[1]?.trim().split(/\s+/)[0]);
+    const text = rows.slice(timingIndex + 1).join(" ");
+    if (start !== null || end !== null || cleanText(text)) {
+      lines.push({ start, end, text });
+    }
+  }
+  return buildTranscript(format, lines);
+}
+function parseLrc(content) {
+  const lines = [];
+  for (const rawLine of content.replace(/\r/g, "").split("\n")) {
+    const matches = [...rawLine.matchAll(/\[(\d{1,3}):(\d{2}(?:\.\d{1,3})?)\]/g)];
+    const text = rawLine.replace(/\[[^\]]+\]/g, "").trim();
+    if (!text || matches.length === 0) continue;
+    for (const match of matches) {
+      const minutes = Number(match[1]);
+      const seconds = Number(match[2]);
+      lines.push({ start: minutes * 60 + seconds, end: null, text });
+    }
+  }
+  lines.sort((a, b) => (a.start ?? 0) - (b.start ?? 0));
+  for (let i = 0; i < lines.length - 1; i += 1) {
+    lines[i].end = lines[i + 1].start;
+  }
+  return buildTranscript("lrc", lines);
+}
+function parseTxt(content) {
+  return buildTranscript(
+    "txt",
+    content.replace(/^\uFEFF/, "").replace(/\r/g, "").split("\n").map((text) => ({ start: null, end: null, text }))
+  );
+}
+function objectValue(obj, keys) {
+  for (const key of keys) {
+    if (key in obj) return obj[key];
+  }
+  return void 0;
+}
+function lineFromUnknown(value) {
+  if (typeof value === "string") {
+    const text2 = cleanText(value);
+    return text2 ? { start: null, end: null, text: text2 } : null;
+  }
+  if (!value || typeof value !== "object" || Array.isArray(value)) return null;
+  const obj = value;
+  const text = cleanText(objectValue(obj, ["text", "phrase", "content", "line", "lyrics", "lyric"]));
+  if (!text) return null;
+  const start = toSeconds(objectValue(obj, ["start", "start_time", "startTime", "begin", "from", "timestamp"]));
+  let end = toSeconds(objectValue(obj, ["end", "end_time", "endTime", "stop", "to"]));
+  if (end === null && start !== null) {
+    const duration = toSeconds(objectValue(obj, ["duration", "length"]));
+    if (duration !== null) end = start + duration;
+  }
+  const type = cleanText(objectValue(obj, ["type", "kind", "segment_type", "segmentType"]));
+  return type ? { start, end, text, type } : { start, end, text };
+}
+function findJsonLineArray(value) {
+  if (Array.isArray(value)) return Array.from(value);
+  if (!value || typeof value !== "object") return null;
+  const obj = value;
+  for (const key of ["phrases", "phrase_segments", "phraseSegments", "segments", "chunks", "lines", "items", "transcript"]) {
+    const candidate = obj[key];
+    if (Array.isArray(candidate)) return Array.from(candidate);
+  }
+  return null;
+}
+function looksLikePhraseJson(value) {
+  if (!value || typeof value !== "object" || Array.isArray(value)) return false;
+  const obj = value;
+  if (["phrases", "phrase_segments", "phraseSegments", "chunks"].some((key) => Array.isArray(obj[key]))) {
+    return true;
+  }
+  const engineTarget = cleanText(obj.engine_version_target).toLowerCase();
+  if (engineTarget.includes("music transcription")) return true;
+  const segments = obj.segments;
+  if (!Array.isArray(segments)) return false;
+  return segments.some((segment) => {
+    if (!segment || typeof segment !== "object" || Array.isArray(segment)) return false;
+    const segmentObj = segment;
+    const type = cleanText(segmentObj.type).toLowerCase();
+    return type === "lyrics" || type === "music";
+  });
+}
+function parseJson(content, requestedFormat) {
+  const parsed = JSON.parse(content);
+  const format = requestedFormat ?? (looksLikePhraseJson(parsed) ? "phrase-json" : "json");
+  const values = findJsonLineArray(parsed);
+  if (!values) {
+    const single = lineFromUnknown(parsed);
+    return buildTranscript(format, single ? [single] : []);
+  }
+  return buildTranscript(
+    format,
+    values.map(lineFromUnknown).filter((line) => line !== null)
+  );
+}
+function detectTranscriptFormat(content, filename) {
+  const lowerName = filename?.toLowerCase() ?? "";
+  if (lowerName.endsWith(".phrase-timing.json") || lowerName.endsWith(".phrase.json") || lowerName.endsWith(".phrases.json")) {
+    return "phrase-json";
+  }
+  if (lowerName.endsWith(".srt")) return "srt";
+  if (lowerName.endsWith(".vtt")) return "vtt";
+  if (lowerName.endsWith(".lrc")) return "lrc";
+  if (lowerName.endsWith(".txt")) return "txt";
+  if (lowerName.endsWith(".json")) {
+    try {
+      const parsed = JSON.parse(content);
+      return looksLikePhraseJson(parsed) ? "phrase-json" : "json";
+    } catch {
+      return "json";
+    }
+  }
+  const trimmed = content.trim();
+  if (/^WEBVTT\b/.test(trimmed)) return "vtt";
+  if (/\d{1,2}:\d{2}:\d{2}[,.]\d{3}\s+-->\s+\d{1,2}:\d{2}:\d{2}[,.]\d{3}/.test(trimmed)) return "srt";
+  if (/^\s*\[\d{1,3}:\d{2}(?:\.\d{1,3})?\]/m.test(trimmed)) return "lrc";
+  if (trimmed.startsWith("{") && trimmed.endsWith("}") || trimmed.startsWith("[") && trimmed.endsWith("]")) {
+    try {
+      const parsed = JSON.parse(trimmed);
+      return looksLikePhraseJson(parsed) ? "phrase-json" : "json";
+    } catch {
+    }
+  }
+  return "txt";
+}
+function parseTranscript(content, options = {}) {
+  const format = options.format ?? detectTranscriptFormat(content, options.filename);
+  switch (format) {
+    case "srt":
+      return parseTimedBlocks(content, "srt");
+    case "vtt":
+      return parseTimedBlocks(content, "vtt");
+    case "lrc":
+      return parseLrc(content);
+    case "txt":
+      return parseTxt(content);
+    case "json":
+      return parseJson(content, "json");
+    case "phrase-json":
+      return parseJson(content, "phrase-json");
+  }
+}
+
+export { detectTranscriptFormat, parseTranscript };
+//# sourceMappingURL=index.js.map
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/index.ts"],"names":["text"],"mappings":";AA0BA,SAAS,UAAU,KAAA,EAAwB;AACzC,EAAA,OAAO,OAAO,UAAU,QAAA,GAAW,KAAA,CAAM,QAAQ,MAAA,EAAQ,GAAG,CAAA,CAAE,IAAA,EAAK,GAAI,EAAA;AACzE;AAEA,SAAS,UAAU,KAAA,EAA+B;AAChD,EAAA,IAAI,OAAO,KAAA,KAAU,QAAA,IAAY,OAAO,QAAA,CAAS,KAAK,GAAG,OAAO,KAAA;AAChE,EAAA,IAAI,OAAO,KAAA,KAAU,QAAA,EAAU,OAAO,IAAA;AAEtC,EAAA,MAAM,aAAa,KAAA,CAAM,IAAA,EAAK,CAAE,OAAA,CAAQ,KAAK,GAAG,CAAA;AAChD,EAAA,IAAI,CAAC,YAAY,OAAO,IAAA;AAExB,EAAA,IAAI,iBAAA,CAAkB,IAAA,CAAK,UAAU,CAAA,EAAG;AACtC,IAAA,MAAM,MAAA,GAAS,OAAO,UAAU,CAAA;AAChC,IAAA,OAAO,MAAA,CAAO,QAAA,CAAS,MAAM,CAAA,GAAI,MAAA,GAAS,IAAA;AAAA,EAC5C;AAEA,EAAA,MAAM,QAAQ,UAAA,CAAW,KAAA,CAAM,GAAG,CAAA,CAAE,IAAI,MAAM,CAAA;AAC9C,EAAA,IAAI,KAAA,CAAM,IAAA,CAAK,CAAC,IAAA,KAAS,CAAC,OAAO,QAAA,CAAS,IAAI,CAAC,CAAA,EAAG,OAAO,IAAA;AAEzD,EAAA,IAAI,KAAA,CAAM,MAAA,KAAW,CAAA,EAAG,OAAO,KAAA,CAAM,CAAC,CAAA,GAAK,IAAA,GAAO,KAAA,CAAM,CAAC,CAAA,GAAK,EAAA,GAAK,MAAM,CAAC,CAAA;AAC1E,EAAA,IAAI,KAAA,CAAM,WAAW,CAAA,EAAG,OAAO,MAAM,CAAC,CAAA,GAAK,EAAA,GAAK,KAAA,CAAM,CAAC,CAAA;AAEvD,EAAA,OAAO,IAAA;AACT;AAEA,SAAS,eAAA,CAAgB,QAA0B,KAAA,EAAqC;AACtF,EAAA,MAAM,QAAA,GAAW,MACd,GAAA,CAAI,CAAC,UAAU,EAAE,GAAG,MAAM,IAAA,EAAM,SAAA,CAAU,KAAK,IAAI,CAAA,GAAI,CAAA,CACvD,MAAA,CAAO,CAAC,IAAA,KAAS,IAAA,CAAK,IAAA,CAAK,MAAA,GAAS,CAAC,CAAA;AAExC,EAAA,OAAO;AAAA,IACL,MAAA;AAAA,IACA,KAAA,EAAO,QAAA;AAAA,IACP,IAAA,EAAM,SAAS,GAAA,CAAI,CAAC,SAAS,IAAA,CAAK,IAAI,CAAA,CAAE,IAAA,CAAK,IAAI;AAAA,GACnD;AACF;AAEA,SAAS,gBAAA,CAAiB,SAAiB,MAAA,EAAmC;AAC5E,EAAA,MAAM,UAAA,GAAa,QAAQ,OAAA,CAAQ,SAAA,EAAW,EAAE,CAAA,CAAE,OAAA,CAAQ,OAAO,EAAE,CAAA;AACnE,EAAA,MAAM,OAAO,MAAA,KAAW,KAAA,GAAQ,WAAW,OAAA,CAAQ,kBAAA,EAAoB,EAAE,CAAA,GAAI,UAAA;AAC7E,EAAA,MAAM,MAAA,GAAS,IAAA,CAAK,KAAA,CAAM,QAAQ,CAAA;AAClC,EAAA,MAAM,QAA0B,EAAC;AAEjC,EAAA,KAAA,MAAW,SAAS,MAAA,EAAQ;AAC1B,IAAA,MAAM,OAAO,KAAA,CAAM,KAAA,CAAM,IAAI,CAAA,CAAE,OAAO,OAAO,CAAA;AAC7C,IAAA,IAAI,IAAA,CAAK,SAAS,CAAA,EAAG;AAErB,IAAA,MAAM,WAAA,GAAc,KAAK,SAAA,CAAU,CAAC,QAAQ,GAAA,CAAI,QAAA,CAAS,KAAK,CAAC,CAAA;AAC/D,IAAA,IAAI,cAAc,CAAA,EAAG;AAErB,IAAA,MAAM,MAAA,GAAS,IAAA,CAAK,WAAW,CAAA,CAAG,MAAM,KAAK,CAAA;AAC7C,IAAA,MAAM,KAAA,GAAQ,SAAA,CAAU,MAAA,CAAO,CAAC,CAAC,CAAA;AACjC,IAAA,MAAM,GAAA,GAAM,SAAA,CAAU,MAAA,CAAO,CAAC,CAAA,EAAG,IAAA,EAAK,CAAE,KAAA,CAAM,KAAK,CAAA,CAAE,CAAC,CAAC,CAAA;AACvD,IAAA,MAAM,OAAO,IAAA,CAAK,KAAA,CAAM,cAAc,CAAC,CAAA,CAAE,KAAK,GAAG,CAAA;AAEjD,IAAA,IAAI,UAAU,IAAA,IAAQ,GAAA,KAAQ,IAAA,IAAQ,SAAA,CAAU,IAAI,CAAA,EAAG;AACrD,MAAA,KAAA,CAAM,IAAA,CAAK,EAAE,KAAA,EAAO,GAAA,EAAK,MAAM,CAAA;AAAA,IACjC;AAAA,EACF;AAEA,EAAA,OAAO,eAAA,CAAgB,QAAQ,KAAK,CAAA;AACtC;AAEA,SAAS,SAAS,OAAA,EAA6B;AAC7C,EAAA,MAAM,QAA0B,EAAC;AAEjC,EAAA,KAAA,MAAW,OAAA,IAAW,QAAQ,OAAA,CAAQ,KAAA,EAAO,EAAE,CAAA,CAAE,KAAA,CAAM,IAAI,CAAA,EAAG;AAC5D,IAAA,MAAM,UAAU,CAAC,GAAG,OAAA,CAAQ,QAAA,CAAS,sCAAsC,CAAC,CAAA;AAC5E,IAAA,MAAM,OAAO,OAAA,CAAQ,OAAA,CAAQ,aAAA,EAAe,EAAE,EAAE,IAAA,EAAK;AACrD,IAAA,IAAI,CAAC,IAAA,IAAQ,OAAA,CAAQ,MAAA,KAAW,CAAA,EAAG;AAEnC,IAAA,KAAA,MAAW,SAAS,OAAA,EAAS;AAC3B,MAAA,MAAM,OAAA,GAAU,MAAA,CAAO,KAAA,CAAM,CAAC,CAAC,CAAA;AAC/B,MAAA,MAAM,OAAA,GAAU,MAAA,CAAO,KAAA,CAAM,CAAC,CAAC,CAAA;AAC/B,MAAA,KAAA,CAAM,IAAA,CAAK,EAAE,KAAA,EAAO,OAAA,GAAU,KAAK,OAAA,EAAS,GAAA,EAAK,IAAA,EAAM,IAAA,EAAM,CAAA;AAAA,IAC/D;AAAA,EACF;AAEA,EAAA,KAAA,CAAM,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAA,CAAO,EAAE,KAAA,IAAS,CAAA,KAAM,CAAA,CAAE,KAAA,IAAS,CAAA,CAAE,CAAA;AACpD,EAAA,KAAA,IAAS,IAAI,CAAA,EAAG,CAAA,GAAI,MAAM,MAAA,GAAS,CAAA,EAAG,KAAK,CAAA,EAAG;AAC5C,IAAA,KAAA,CAAM,CAAC,CAAA,CAAG,GAAA,GAAM,KAAA,CAAM,CAAA,GAAI,CAAC,CAAA,CAAG,KAAA;AAAA,EAChC;AAEA,EAAA,OAAO,eAAA,CAAgB,OAAO,KAAK,CAAA;AACrC;AAEA,SAAS,SAAS,OAAA,EAA6B;AAC7C,EAAA,OAAO,eAAA;AAAA,IACL,KAAA;AAAA,IACA,OAAA,CACG,QAAQ,SAAA,EAAW,EAAE,EACrB,OAAA,CAAQ,KAAA,EAAO,EAAE,CAAA,CACjB,KAAA,CAAM,IAAI,CAAA,CACV,GAAA,CAAI,CAAC,IAAA,MAAU,EAAE,OAAO,IAAA,EAAM,GAAA,EAAK,IAAA,EAAM,IAAA,EAAK,CAAE;AAAA,GACrD;AACF;AAEA,SAAS,WAAA,CAAY,KAA8B,IAAA,EAAyB;AAC1E,EAAA,KAAA,MAAW,OAAO,IAAA,EAAM;AACtB,IAAA,IAAI,GAAA,IAAO,GAAA,EAAK,OAAO,GAAA,CAAI,GAAG,CAAA;AAAA,EAChC;AACA,EAAA,OAAO,MAAA;AACT;AAEA,SAAS,gBAAgB,KAAA,EAAuC;AAC9D,EAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,IAAA,MAAMA,KAAAA,GAAO,UAAU,KAAK,CAAA;AAC5B,IAAA,OAAOA,KAAAA,GAAO,EAAE,KAAA,EAAO,IAAA,EAAM,KAAK,IAAA,EAAM,IAAA,EAAAA,OAAK,GAAI,IAAA;AAAA,EACnD;AAEA,EAAA,IAAI,CAAC,SAAS,OAAO,KAAA,KAAU,YAAY,KAAA,CAAM,OAAA,CAAQ,KAAK,CAAA,EAAG,OAAO,IAAA;AACxE,EAAA,MAAM,GAAA,GAAM,KAAA;AACZ,EAAA,MAAM,IAAA,GAAO,SAAA,CAAU,WAAA,CAAY,GAAA,EAAK,CAAC,MAAA,EAAQ,QAAA,EAAU,SAAA,EAAW,MAAA,EAAQ,QAAA,EAAU,OAAO,CAAC,CAAC,CAAA;AACjG,EAAA,IAAI,CAAC,MAAM,OAAO,IAAA;AAElB,EAAA,MAAM,KAAA,GAAQ,SAAA,CAAU,WAAA,CAAY,GAAA,EAAK,CAAC,OAAA,EAAS,YAAA,EAAc,WAAA,EAAa,OAAA,EAAS,MAAA,EAAQ,WAAW,CAAC,CAAC,CAAA;AAC5G,EAAA,IAAI,GAAA,GAAM,SAAA,CAAU,WAAA,CAAY,GAAA,EAAK,CAAC,KAAA,EAAO,UAAA,EAAY,SAAA,EAAW,MAAA,EAAQ,IAAI,CAAC,CAAC,CAAA;AAElF,EAAA,IAAI,GAAA,KAAQ,IAAA,IAAQ,KAAA,KAAU,IAAA,EAAM;AAClC,IAAA,MAAM,QAAA,GAAW,UAAU,WAAA,CAAY,GAAA,EAAK,CAAC,UAAA,EAAY,QAAQ,CAAC,CAAC,CAAA;AACnE,IAAA,IAAI,QAAA,KAAa,IAAA,EAAM,GAAA,GAAM,KAAA,GAAQ,QAAA;AAAA,EACvC;AAEA,EAAA,MAAM,IAAA,GAAO,SAAA,CAAU,WAAA,CAAY,GAAA,EAAK,CAAC,QAAQ,MAAA,EAAQ,cAAA,EAAgB,aAAa,CAAC,CAAC,CAAA;AACxF,EAAA,OAAO,IAAA,GAAO,EAAE,KAAA,EAAO,GAAA,EAAK,IAAA,EAAM,MAAK,GAAI,EAAE,KAAA,EAAO,GAAA,EAAK,IAAA,EAAK;AAChE;AAEA,SAAS,kBAAkB,KAAA,EAAkC;AAC3D,EAAA,IAAI,MAAM,OAAA,CAAQ,KAAK,GAAG,OAAO,KAAA,CAAM,KAAK,KAAkB,CAAA;AAC9D,EAAA,IAAI,CAAC,KAAA,IAAS,OAAO,KAAA,KAAU,UAAU,OAAO,IAAA;AAEhD,EAAA,MAAM,GAAA,GAAM,KAAA;AACZ,EAAA,KAAA,MAAW,GAAA,IAAO,CAAC,SAAA,EAAW,iBAAA,EAAmB,gBAAA,EAAkB,YAAY,QAAA,EAAU,OAAA,EAAS,OAAA,EAAS,YAAY,CAAA,EAAG;AACxH,IAAA,MAAM,SAAA,GAAY,IAAI,GAAG,CAAA;AACzB,IAAA,IAAI,MAAM,OAAA,CAAQ,SAAS,GAAG,OAAO,KAAA,CAAM,KAAK,SAAsB,CAAA;AAAA,EACxE;AAEA,EAAA,OAAO,IAAA;AACT;AAEA,SAAS,oBAAoB,KAAA,EAAyB;AACpD,EAAA,IAAI,CAAC,SAAS,OAAO,KAAA,KAAU,YAAY,KAAA,CAAM,OAAA,CAAQ,KAAK,CAAA,EAAG,OAAO,KAAA;AACxE,EAAA,MAAM,GAAA,GAAM,KAAA;AAEZ,EAAA,IAAI,CAAC,SAAA,EAAW,iBAAA,EAAmB,gBAAA,EAAkB,QAAQ,CAAA,CAAE,IAAA,CAAK,CAAC,GAAA,KAAQ,MAAM,OAAA,CAAQ,GAAA,CAAI,GAAG,CAAC,CAAC,CAAA,EAAG;AACrG,IAAA,OAAO,IAAA;AAAA,EACT;AAEA,EAAA,MAAM,YAAA,GAAe,SAAA,CAAU,GAAA,CAAI,qBAAqB,EAAE,WAAA,EAAY;AACtE,EAAA,IAAI,YAAA,CAAa,QAAA,CAAS,qBAAqB,CAAA,EAAG,OAAO,IAAA;AAEzD,EAAA,MAAM,WAAW,GAAA,CAAI,QAAA;AACrB,EAAA,IAAI,CAAC,KAAA,CAAM,OAAA,CAAQ,QAAQ,GAAG,OAAO,KAAA;AAErC,EAAA,OAAO,QAAA,CAAS,IAAA,CAAK,CAAC,OAAA,KAAqB;AACzC,IAAA,IAAI,CAAC,WAAW,OAAO,OAAA,KAAY,YAAY,KAAA,CAAM,OAAA,CAAQ,OAAO,CAAA,EAAG,OAAO,KAAA;AAC9E,IAAA,MAAM,UAAA,GAAa,OAAA;AACnB,IAAA,MAAM,IAAA,GAAO,SAAA,CAAU,UAAA,CAAW,IAAI,EAAE,WAAA,EAAY;AACpD,IAAA,OAAO,IAAA,KAAS,YAAY,IAAA,KAAS,OAAA;AAAA,EACvC,CAAC,CAAA;AACH;AAEA,SAAS,SAAA,CAAU,SAAiB,eAAA,EAAsD;AACxF,EAAA,MAAM,MAAA,GAAkB,IAAA,CAAK,KAAA,CAAM,OAAO,CAAA;AAC1C,EAAA,MAAM,MAAA,GAAiC,eAAA,KAAoB,mBAAA,CAAoB,MAAM,IAAI,aAAA,GAAgB,MAAA,CAAA;AACzG,EAAA,MAAM,MAAA,GAAS,kBAAkB,MAAM,CAAA;AAEvC,EAAA,IAAI,CAAC,MAAA,EAAQ;AACX,IAAA,MAAM,MAAA,GAAS,gBAAgB,MAAM,CAAA;AACrC,IAAA,OAAO,gBAAgB,MAAA,EAAQ,MAAA,GAAS,CAAC,MAAM,CAAA,GAAI,EAAE,CAAA;AAAA,EACvD;AAEA,EAAA,OAAO,eAAA;AAAA,IACL,MAAA;AAAA,IACA,MAAA,CAAO,IAAI,eAAe,CAAA,CAAE,OAAO,CAAC,IAAA,KAAiC,SAAS,IAAI;AAAA,GACpF;AACF;AAEO,SAAS,sBAAA,CAAuB,SAAiB,QAAA,EAAqC;AAC3F,EAAA,MAAM,SAAA,GAAY,QAAA,EAAU,WAAA,EAAY,IAAK,EAAA;AAE7C,EAAA,IACE,SAAA,CAAU,QAAA,CAAS,qBAAqB,CAAA,IACxC,SAAA,CAAU,QAAA,CAAS,cAAc,CAAA,IACjC,SAAA,CAAU,QAAA,CAAS,eAAe,CAAA,EAClC;AACA,IAAA,OAAO,aAAA;AAAA,EACT;AACA,EAAA,IAAI,SAAA,CAAU,QAAA,CAAS,MAAM,CAAA,EAAG,OAAO,KAAA;AACvC,EAAA,IAAI,SAAA,CAAU,QAAA,CAAS,MAAM,CAAA,EAAG,OAAO,KAAA;AACvC,EAAA,IAAI,SAAA,CAAU,QAAA,CAAS,MAAM,CAAA,EAAG,OAAO,KAAA;AACvC,EAAA,IAAI,SAAA,CAAU,QAAA,CAAS,MAAM,CAAA,EAAG,OAAO,KAAA;AACvC,EAAA,IAAI,SAAA,CAAU,QAAA,CAAS,OAAO,CAAA,EAAG;AAC/B,IAAA,IAAI;AACF,MAAA,MAAM,MAAA,GAAkB,IAAA,CAAK,KAAA,CAAM,OAAO,CAAA;AAC1C,MAAA,OAAO,mBAAA,CAAoB,MAAM,CAAA,GAAI,aAAA,GAAgB,MAAA;AAAA,IACvD,CAAA,CAAA,MAAQ;AACN,MAAA,OAAO,MAAA;AAAA,IACT;AAAA,EACF;AAEA,EAAA,MAAM,OAAA,GAAU,QAAQ,IAAA,EAAK;AAC7B,EAAA,IAAI,WAAA,CAAY,IAAA,CAAK,OAAO,CAAA,EAAG,OAAO,KAAA;AACtC,EAAA,IAAI,mEAAA,CAAoE,IAAA,CAAK,OAAO,CAAA,EAAG,OAAO,KAAA;AAC9F,EAAA,IAAI,sCAAA,CAAuC,IAAA,CAAK,OAAO,CAAA,EAAG,OAAO,KAAA;AAEjE,EAAA,IAAK,OAAA,CAAQ,UAAA,CAAW,GAAG,CAAA,IAAK,QAAQ,QAAA,CAAS,GAAG,CAAA,IAAO,OAAA,CAAQ,WAAW,GAAG,CAAA,IAAK,OAAA,CAAQ,QAAA,CAAS,GAAG,CAAA,EAAI;AAC5G,IAAA,IAAI;AACF,MAAA,MAAM,MAAA,GAAkB,IAAA,CAAK,KAAA,CAAM,OAAO,CAAA;AAC1C,MAAA,OAAO,mBAAA,CAAoB,MAAM,CAAA,GAAI,aAAA,GAAgB,MAAA;AAAA,IACvD,CAAA,CAAA,MAAQ;AAAA,IAER;AAAA,EACF;AAEA,EAAA,OAAO,KAAA;AACT;AAEO,SAAS,eAAA,CAAgB,OAAA,EAAiB,OAAA,GAAkC,EAAC,EAAe;AACjG,EAAA,MAAM,SAAS,OAAA,CAAQ,MAAA,IAAU,sBAAA,CAAuB,OAAA,EAAS,QAAQ,QAAQ,CAAA;AAEjF,EAAA,QAAQ,MAAA;AAAQ,IACd,KAAK,KAAA;AACH,MAAA,OAAO,gBAAA,CAAiB,SAAS,KAAK,CAAA;AAAA,IACxC,KAAK,KAAA;AACH,MAAA,OAAO,gBAAA,CAAiB,SAAS,KAAK,CAAA;AAAA,IACxC,KAAK,KAAA;AACH,MAAA,OAAO,SAAS,OAAO,CAAA;AAAA,IACzB,KAAK,KAAA;AACH,MAAA,OAAO,SAAS,OAAO,CAAA;AAAA,IACzB,KAAK,MAAA;AACH,MAAA,OAAO,SAAA,CAAU,SAAS,MAAM,CAAA;AAAA,IAClC,KAAK,aAAA;AACH,MAAA,OAAO,SAAA,CAAU,SAAS,aAAa,CAAA;AAAA;AAE7C","file":"index.js","sourcesContent":["export type TranscriptFormat =\n  | 'srt'\n  | 'vtt'\n  | 'lrc'\n  | 'txt'\n  | 'json'\n  | 'phrase-json';\n\nexport interface TranscriptLine {\n  start: number | null;\n  end: number | null;\n  text: string;\n  type?: string;\n}\n\nexport interface Transcript {\n  format: TranscriptFormat;\n  lines: TranscriptLine[];\n  text: string;\n}\n\nexport interface ParseTranscriptOptions {\n  format?: TranscriptFormat;\n  filename?: string;\n}\n\nfunction cleanText(value: unknown): string {\n  return typeof value === 'string' ? value.replace(/\\s+/g, ' ').trim() : '';\n}\n\nfunction toSeconds(value: unknown): number | null {\n  if (typeof value === 'number' && Number.isFinite(value)) return value;\n  if (typeof value !== 'string') return null;\n\n  const normalized = value.trim().replace(',', '.');\n  if (!normalized) return null;\n\n  if (/^\\d+(?:\\.\\d+)?$/.test(normalized)) {\n    const parsed = Number(normalized);\n    return Number.isFinite(parsed) ? parsed : null;\n  }\n\n  const parts = normalized.split(':').map(Number);\n  if (parts.some((part) => !Number.isFinite(part))) return null;\n\n  if (parts.length === 3) return parts[0]! * 3600 + parts[1]! * 60 + parts[2]!;\n  if (parts.length === 2) return parts[0]! * 60 + parts[1]!;\n\n  return null;\n}\n\nfunction buildTranscript(format: TranscriptFormat, lines: TranscriptLine[]): Transcript {\n  const filtered = lines\n    .map((line) => ({ ...line, text: cleanText(line.text) }))\n    .filter((line) => line.text.length > 0);\n\n  return {\n    format,\n    lines: filtered,\n    text: filtered.map((line) => line.text).join('\\n')\n  };\n}\n\nfunction parseTimedBlocks(content: string, format: 'srt' | 'vtt'): Transcript {\n  const normalized = content.replace(/^\\uFEFF/, '').replace(/\\r/g, '');\n  const body = format === 'vtt' ? normalized.replace(/^WEBVTT[^\\n]*\\n+/, '') : normalized;\n  const blocks = body.split(/\\n{2,}/);\n  const lines: TranscriptLine[] = [];\n\n  for (const block of blocks) {\n    const rows = block.split('\\n').filter(Boolean);\n    if (rows.length < 2) continue;\n\n    const timingIndex = rows.findIndex((row) => row.includes('-->'));\n    if (timingIndex < 0) continue;\n\n    const timing = rows[timingIndex]!.split('-->');\n    const start = toSeconds(timing[0]);\n    const end = toSeconds(timing[1]?.trim().split(/\\s+/)[0]);\n    const text = rows.slice(timingIndex + 1).join(' ');\n\n    if (start !== null || end !== null || cleanText(text)) {\n      lines.push({ start, end, text });\n    }\n  }\n\n  return buildTranscript(format, lines);\n}\n\nfunction parseLrc(content: string): Transcript {\n  const lines: TranscriptLine[] = [];\n\n  for (const rawLine of content.replace(/\\r/g, '').split('\\n')) {\n    const matches = [...rawLine.matchAll(/\\[(\\d{1,3}):(\\d{2}(?:\\.\\d{1,3})?)\\]/g)];\n    const text = rawLine.replace(/\\[[^\\]]+\\]/g, '').trim();\n    if (!text || matches.length === 0) continue;\n\n    for (const match of matches) {\n      const minutes = Number(match[1]);\n      const seconds = Number(match[2]);\n      lines.push({ start: minutes * 60 + seconds, end: null, text });\n    }\n  }\n\n  lines.sort((a, b) => (a.start ?? 0) - (b.start ?? 0));\n  for (let i = 0; i < lines.length - 1; i += 1) {\n    lines[i]!.end = lines[i + 1]!.start;\n  }\n\n  return buildTranscript('lrc', lines);\n}\n\nfunction parseTxt(content: string): Transcript {\n  return buildTranscript(\n    'txt',\n    content\n      .replace(/^\\uFEFF/, '')\n      .replace(/\\r/g, '')\n      .split('\\n')\n      .map((text) => ({ start: null, end: null, text }))\n  );\n}\n\nfunction objectValue(obj: Record<string, unknown>, keys: string[]): unknown {\n  for (const key of keys) {\n    if (key in obj) return obj[key];\n  }\n  return undefined;\n}\n\nfunction lineFromUnknown(value: unknown): TranscriptLine | null {\n  if (typeof value === 'string') {\n    const text = cleanText(value);\n    return text ? { start: null, end: null, text } : null;\n  }\n\n  if (!value || typeof value !== 'object' || Array.isArray(value)) return null;\n  const obj = value as Record<string, unknown>;\n  const text = cleanText(objectValue(obj, ['text', 'phrase', 'content', 'line', 'lyrics', 'lyric']));\n  if (!text) return null;\n\n  const start = toSeconds(objectValue(obj, ['start', 'start_time', 'startTime', 'begin', 'from', 'timestamp']));\n  let end = toSeconds(objectValue(obj, ['end', 'end_time', 'endTime', 'stop', 'to']));\n\n  if (end === null && start !== null) {\n    const duration = toSeconds(objectValue(obj, ['duration', 'length']));\n    if (duration !== null) end = start + duration;\n  }\n\n  const type = cleanText(objectValue(obj, ['type', 'kind', 'segment_type', 'segmentType']));\n  return type ? { start, end, text, type } : { start, end, text };\n}\n\nfunction findJsonLineArray(value: unknown): unknown[] | null {\n  if (Array.isArray(value)) return Array.from(value as unknown[]);\n  if (!value || typeof value !== 'object') return null;\n\n  const obj = value as Record<string, unknown>;\n  for (const key of ['phrases', 'phrase_segments', 'phraseSegments', 'segments', 'chunks', 'lines', 'items', 'transcript']) {\n    const candidate = obj[key];\n    if (Array.isArray(candidate)) return Array.from(candidate as unknown[]);\n  }\n\n  return null;\n}\n\nfunction looksLikePhraseJson(value: unknown): boolean {\n  if (!value || typeof value !== 'object' || Array.isArray(value)) return false;\n  const obj = value as Record<string, unknown>;\n\n  if (['phrases', 'phrase_segments', 'phraseSegments', 'chunks'].some((key) => Array.isArray(obj[key]))) {\n    return true;\n  }\n\n  const engineTarget = cleanText(obj.engine_version_target).toLowerCase();\n  if (engineTarget.includes('music transcription')) return true;\n\n  const segments = obj.segments;\n  if (!Array.isArray(segments)) return false;\n\n  return segments.some((segment: unknown) => {\n    if (!segment || typeof segment !== 'object' || Array.isArray(segment)) return false;\n    const segmentObj = segment as Record<string, unknown>;\n    const type = cleanText(segmentObj.type).toLowerCase();\n    return type === 'lyrics' || type === 'music';\n  });\n}\n\nfunction parseJson(content: string, requestedFormat?: 'json' | 'phrase-json'): Transcript {\n  const parsed: unknown = JSON.parse(content);\n  const format: 'json' | 'phrase-json' = requestedFormat ?? (looksLikePhraseJson(parsed) ? 'phrase-json' : 'json');\n  const values = findJsonLineArray(parsed);\n\n  if (!values) {\n    const single = lineFromUnknown(parsed);\n    return buildTranscript(format, single ? [single] : []);\n  }\n\n  return buildTranscript(\n    format,\n    values.map(lineFromUnknown).filter((line): line is TranscriptLine => line !== null)\n  );\n}\n\nexport function detectTranscriptFormat(content: string, filename?: string): TranscriptFormat {\n  const lowerName = filename?.toLowerCase() ?? '';\n\n  if (\n    lowerName.endsWith('.phrase-timing.json') ||\n    lowerName.endsWith('.phrase.json') ||\n    lowerName.endsWith('.phrases.json')\n  ) {\n    return 'phrase-json';\n  }\n  if (lowerName.endsWith('.srt')) return 'srt';\n  if (lowerName.endsWith('.vtt')) return 'vtt';\n  if (lowerName.endsWith('.lrc')) return 'lrc';\n  if (lowerName.endsWith('.txt')) return 'txt';\n  if (lowerName.endsWith('.json')) {\n    try {\n      const parsed: unknown = JSON.parse(content);\n      return looksLikePhraseJson(parsed) ? 'phrase-json' : 'json';\n    } catch {\n      return 'json';\n    }\n  }\n\n  const trimmed = content.trim();\n  if (/^WEBVTT\\b/.test(trimmed)) return 'vtt';\n  if (/\\d{1,2}:\\d{2}:\\d{2}[,.]\\d{3}\\s+-->\\s+\\d{1,2}:\\d{2}:\\d{2}[,.]\\d{3}/.test(trimmed)) return 'srt';\n  if (/^\\s*\\[\\d{1,3}:\\d{2}(?:\\.\\d{1,3})?\\]/m.test(trimmed)) return 'lrc';\n\n  if ((trimmed.startsWith('{') && trimmed.endsWith('}')) || (trimmed.startsWith('[') && trimmed.endsWith(']'))) {\n    try {\n      const parsed: unknown = JSON.parse(trimmed);\n      return looksLikePhraseJson(parsed) ? 'phrase-json' : 'json';\n    } catch {\n      // fall through to txt\n    }\n  }\n\n  return 'txt';\n}\n\nexport function parseTranscript(content: string, options: ParseTranscriptOptions = {}): Transcript {\n  const format = options.format ?? detectTranscriptFormat(content, options.filename);\n\n  switch (format) {\n    case 'srt':\n      return parseTimedBlocks(content, 'srt');\n    case 'vtt':\n      return parseTimedBlocks(content, 'vtt');\n    case 'lrc':\n      return parseLrc(content);\n    case 'txt':\n      return parseTxt(content);\n    case 'json':\n      return parseJson(content, 'json');\n    case 'phrase-json':\n      return parseJson(content, 'phrase-json');\n  }\n}\n"]}

package/package.json

@@ -0,0 +1,62 @@
+{
+  "name": "@leelaing/transcripts",
+  "version": "0.1.0",
+  "description": "Transcript and song-lyric parsing utilities for SRT, VTT, LRC, TXT, JSON, and Whisper Music phrase JSON.",
+  "type": "module",
+  "main": "./dist/index.cjs",
+  "module": "./dist/index.js",
+  "types": "./dist/index.d.ts",
+  "exports": {
+    ".": {
+      "types": "./dist/index.d.ts",
+      "import": "./dist/index.js",
+      "require": "./dist/index.cjs"
+    }
+  },
+  "files": [
+    "dist"
+  ],
+  "scripts": {
+    "start": "tsup --watch",
+    "dev": "npm start",
+    "build": "tsup",
+    "test": "vitest run",
+    "test:watch": "vitest",
+    "lint": "eslint .",
+    "format": "prettier --write .",
+    "format:check": "prettier --check .",
+    "typecheck": "tsc --noEmit",
+    "check": "npm run typecheck && npm run lint && npm test && npm run build",
+    "clean": "rm -rf dist coverage",
+    "init-package": "node scripts/init-package.mjs",
+    "prepublishOnly": "npm run check"
+  },
+  "engines": {
+    "node": ">=20"
+  },
+  "keywords": [],
+  "author": "Lee Charles Laing",
+  "license": "MIT",
+  "publishConfig": {
+    "access": "public"
+  },
+  "devDependencies": {
+    "@eslint/js": "10.0.1",
+    "@vitest/coverage-v8": "4.1.11",
+    "eslint": "10.9.1",
+    "globals": "17.11.0",
+    "prettier": "3.9.6",
+    "tsup": "8.5.1",
+    "typescript": "6.0.3",
+    "typescript-eslint": "8.68.0",
+    "vitest": "4.1.11"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/lcharleslaing/lee-transcripts.git"
+  },
+  "bugs": {
+    "url": "https://github.com/lcharleslaing/lee-transcripts/issues"
+  },
+  "homepage": "https://github.com/lcharleslaing/lee-transcripts#readme"
+}