@storyteller-platform/align 0.1.16 → 0.1.18

package/dist/process/ranges.cjs

@@ -75,7 +75,7 @@ var import_vad = require("@storyteller-platform/ghost-story/vad");
 var import_ffmpeg = require("../common/ffmpeg.cjs");
 async function getSafeChapterRanges(input, duration, chapters, maxSeconds, signal, logger) {
   if (!chapters.length) {
-    logger == null ? void 0 : logger.info(
+    logger?.info(
       `Track is longer than ${maxSeconds / 60} minutes (${duration / 60}m); using VAD to determine safe split points.`
     );
     const ranges2 = await getSafeRanges(input, duration, maxSeconds, 0, signal);
@@ -103,7 +103,7 @@ async function getSafeChapterRanges(input, duration, chapters, maxSeconds, signa
       ranges.push(range);
       continue;
     }
-    logger == null ? void 0 : logger.info(
+    logger?.info(
       `Chapter is longer than ${maxSeconds / 60} minutes (${duration / 60}m); using VAD to determine safe split points.`
     );
     const chapterRanges = await getSafeRanges(
@@ -132,7 +132,7 @@ async function getSafeRanges(input, duration, maxSeconds, start = 0, signal, log
     });
     const ranges = [{ start, end: duration + start }];
     for (let i = 0; i + 1 < duration / maxSeconds; i++) {
-      if (signal == null ? void 0 : signal.aborted) throw new Error("Aborted");
+      if (signal?.aborted) throw new Error("Aborted");
       const tmpFilepath = (0, import_node_path.join)(tmpDir, i.toString(), `${rawFilename}.wav`);
       await (0, import_promises.mkdir)((0, import_node_path.dirname)(tmpFilepath), { recursive: true });
       const approxCutPoint = start + maxSeconds * (i + 1);

package/dist/process/ranges.js

@@ -10,7 +10,7 @@ import { detectVoiceActivity } from "@storyteller-platform/ghost-story/vad";
 import { splitFile } from "../common/ffmpeg.js";
 async function getSafeChapterRanges(input, duration, chapters, maxSeconds, signal, logger) {
   if (!chapters.length) {
-    logger == null ? void 0 : logger.info(
+    logger?.info(
       `Track is longer than ${maxSeconds / 60} minutes (${duration / 60}m); using VAD to determine safe split points.`
     );
     const ranges2 = await getSafeRanges(input, duration, maxSeconds, 0, signal);
@@ -38,7 +38,7 @@ async function getSafeChapterRanges(input, duration, chapters, maxSeconds, signa
       ranges.push(range);
       continue;
     }
-    logger == null ? void 0 : logger.info(
+    logger?.info(
       `Chapter is longer than ${maxSeconds / 60} minutes (${duration / 60}m); using VAD to determine safe split points.`
     );
     const chapterRanges = await getSafeRanges(
@@ -67,7 +67,7 @@ async function getSafeRanges(input, duration, maxSeconds, start = 0, signal, log
     });
     const ranges = [{ start, end: duration + start }];
     for (let i = 0; i + 1 < duration / maxSeconds; i++) {
-      if (signal == null ? void 0 : signal.aborted) throw new Error("Aborted");
+      if (signal?.aborted) throw new Error("Aborted");
       const tmpFilepath = join(tmpDir, i.toString(), `${rawFilename}.wav`);
       await mkdir(dirname(tmpFilepath), { recursive: true });
       const approxCutPoint = start + maxSeconds * (i + 1);

package/dist/transcribe/transcribe.cjs

@@ -84,7 +84,6 @@ var import_async_semaphore = require("@esfx/async-semaphore");
 var import_audiobook = require("@storyteller-platform/audiobook");
 var import_ghost_story = require("@storyteller-platform/ghost-story");
 async function transcribe(input, output, locale, options) {
-  var _a;
   if (process.env["DEBUG_TRANSCRIBE"] === "true") {
     const inspector = await import("node:inspector");
     inspector.open(9231, "0.0.0.0", true);
@@ -109,7 +108,7 @@ async function transcribe(input, output, locale, options) {
     await (0, import_ghost_story.ensureWhisperInstalled)({
       model,
       printOutput: ["debug", "info"].includes(
-        ((_a = options.logger) == null ? void 0 : _a.level) ?? "silent"
+        options.logger?.level ?? "silent"
       ),
       signal
     });
@@ -126,7 +125,6 @@ async function transcribe(input, output, locale, options) {
   timing.setMetadata("threads", options.threads ?? 4);
   await Promise.all(
     filenames.map(async (filename) => {
-      var _a2, _b, _c;
       var _stack = [];
       try {
         if (aborted()) throw new Error("Aborted");
@@ -140,7 +138,7 @@ async function transcribe(input, output, locale, options) {
             encoding: "utf-8",
             signal
           });
-          (_a2 = options.logger) == null ? void 0 : _a2.info(`Found existing transcription for ${filepath}`);
+          options.logger?.info(`Found existing transcription for ${filepath}`);
           transcriptions.push(transcriptionFilepath);
         } catch {
         }
@@ -151,13 +149,12 @@ async function transcribe(input, output, locale, options) {
         });
         await semaphore.wait();
         function onFileProgress(progress) {
-          var _a3, _b2;
           perFileProgress.set(filename, progress);
           const updatedProgress = Array.from(perFileProgress.values()).reduce((acc, p) => acc + p) / filenames.length;
-          (_a3 = options.logger) == null ? void 0 : _a3.info(
+          options.logger?.info(
             `Progress: ${Math.floor(updatedProgress * 100)}%`
           );
-          (_b2 = options.onProgress) == null ? void 0 : _b2.call(options, updatedProgress);
+          options.onProgress?.(updatedProgress);
         }
         const transcription = await transcribeFile(filepath, locale, {
           ...options,
@@ -168,7 +165,7 @@ async function transcribe(input, output, locale, options) {
           threads: options.threads ?? 4,
           onProgress: onFileProgress
         });
-        (_b = options.logger) == null ? void 0 : _b.info(
+        options.logger?.info(
           (0, import_ghost_story.formatSingleReport)(
             transcription.timing,
             `Transcription Timing Report for ${filepath}`
@@ -184,7 +181,7 @@ async function transcribe(input, output, locale, options) {
           { signal }
         );
         transcriptions.push(transcriptionFilepath);
-        (_c = options.onProgress) == null ? void 0 : _c.call(options, (transcriptions.length + 1) / filenames.length);
+        options.onProgress?.((transcriptions.length + 1) / filenames.length);
       } catch (_) {
         var _error = _, _hasError = true;
       } finally {
@@ -200,7 +197,6 @@ async function transcribe(input, output, locale, options) {
   return timing;
 }
 async function transcribeFile(input, locale, options) {
-  var _a, _b;
   const audioFilepath = (0, import_node_path.resolve)(process.cwd(), input);
   const sharedOptions = {
     signal: options.signal,
@@ -215,11 +211,11 @@ async function transcribeFile(input, locale, options) {
         model: options.model,
         variant: fallbackVariant,
         printOutput: ["debug", "info"].includes(
-          ((_a = options.logger) == null ? void 0 : _a.level) ?? "silent"
+          options.logger?.level ?? "silent"
         ),
         signal: options.signal
       });
-      (_b = options.logger) == null ? void 0 : _b.info(`Transcribing audio file ${audioFilepath}`);
+      options.logger?.info(`Transcribing audio file ${audioFilepath}`);
       return (0, import_ghost_story.recognize)(audioFilepath, {
         engine: options.engine,
         options: {
@@ -228,12 +224,11 @@ async function transcribeFile(input, locale, options) {
           processors: options.processors,
           threads: options.threads,
           onProgress: (progress) => {
-            var _a2;
             if (options.onProgress) {
               options.onProgress(progress);
               return;
             }
-            (_a2 = options.logger) == null ? void 0 : _a2.info(
+            options.logger?.info(
               `Transcribing ${audioFilepath} progress: ${Math.floor(progress * 100)}%`
             );
           },

package/dist/transcribe/transcribe.js

@@ -15,7 +15,6 @@ import {
   recognize
 } from "@storyteller-platform/ghost-story";
 async function transcribe(input, output, locale, options) {
-  var _a;
   if (process.env["DEBUG_TRANSCRIBE"] === "true") {
     const inspector = await import("node:inspector");
     inspector.open(9231, "0.0.0.0", true);
@@ -40,7 +39,7 @@ async function transcribe(input, output, locale, options) {
     await ensureWhisperInstalled({
       model,
       printOutput: ["debug", "info"].includes(
-        ((_a = options.logger) == null ? void 0 : _a.level) ?? "silent"
+        options.logger?.level ?? "silent"
       ),
       signal
     });
@@ -57,7 +56,6 @@ async function transcribe(input, output, locale, options) {
   timing.setMetadata("threads", options.threads ?? 4);
   await Promise.all(
     filenames.map(async (filename) => {
-      var _a2, _b, _c;
       var _stack = [];
       try {
         if (aborted()) throw new Error("Aborted");
@@ -71,7 +69,7 @@ async function transcribe(input, output, locale, options) {
             encoding: "utf-8",
             signal
           });
-          (_a2 = options.logger) == null ? void 0 : _a2.info(`Found existing transcription for ${filepath}`);
+          options.logger?.info(`Found existing transcription for ${filepath}`);
           transcriptions.push(transcriptionFilepath);
         } catch {
         }
@@ -82,13 +80,12 @@ async function transcribe(input, output, locale, options) {
         });
         await semaphore.wait();
         function onFileProgress(progress) {
-          var _a3, _b2;
           perFileProgress.set(filename, progress);
           const updatedProgress = Array.from(perFileProgress.values()).reduce((acc, p) => acc + p) / filenames.length;
-          (_a3 = options.logger) == null ? void 0 : _a3.info(
+          options.logger?.info(
             `Progress: ${Math.floor(updatedProgress * 100)}%`
           );
-          (_b2 = options.onProgress) == null ? void 0 : _b2.call(options, updatedProgress);
+          options.onProgress?.(updatedProgress);
         }
         const transcription = await transcribeFile(filepath, locale, {
           ...options,
@@ -99,7 +96,7 @@ async function transcribe(input, output, locale, options) {
           threads: options.threads ?? 4,
           onProgress: onFileProgress
         });
-        (_b = options.logger) == null ? void 0 : _b.info(
+        options.logger?.info(
           formatSingleReport(
             transcription.timing,
             `Transcription Timing Report for ${filepath}`
@@ -115,7 +112,7 @@ async function transcribe(input, output, locale, options) {
           { signal }
         );
         transcriptions.push(transcriptionFilepath);
-        (_c = options.onProgress) == null ? void 0 : _c.call(options, (transcriptions.length + 1) / filenames.length);
+        options.onProgress?.((transcriptions.length + 1) / filenames.length);
       } catch (_) {
         var _error = _, _hasError = true;
       } finally {
@@ -131,7 +128,6 @@ async function transcribe(input, output, locale, options) {
   return timing;
 }
 async function transcribeFile(input, locale, options) {
-  var _a, _b;
   const audioFilepath = resolve(process.cwd(), input);
   const sharedOptions = {
     signal: options.signal,
@@ -146,11 +142,11 @@ async function transcribeFile(input, locale, options) {
         model: options.model,
         variant: fallbackVariant,
         printOutput: ["debug", "info"].includes(
-          ((_a = options.logger) == null ? void 0 : _a.level) ?? "silent"
+          options.logger?.level ?? "silent"
         ),
         signal: options.signal
       });
-      (_b = options.logger) == null ? void 0 : _b.info(`Transcribing audio file ${audioFilepath}`);
+      options.logger?.info(`Transcribing audio file ${audioFilepath}`);
       return recognize(audioFilepath, {
         engine: options.engine,
         options: {
@@ -159,12 +155,11 @@ async function transcribeFile(input, locale, options) {
           processors: options.processors,
           threads: options.threads,
           onProgress: (progress) => {
-            var _a2;
             if (options.onProgress) {
               options.onProgress(progress);
               return;
             }
-            (_a2 = options.logger) == null ? void 0 : _a2.info(
+            options.logger?.info(
               `Transcribing ${audioFilepath} progress: ${Math.floor(progress * 100)}%`
             );
           },

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@storyteller-platform/align",
-  "version": "0.1.16",
+  "version": "0.1.18",
   "description": "A library and CLI for automatically aligning audiobooks and EPUBs to produce Media Overlays",
   "author": "Shane Friedman",
   "license": "MIT",
@@ -28,6 +28,8 @@
   "type": "module",
   "files": [
     "dist",
+    "prebuilds",
+    "binding.gyp",
     "README.md",
     "LICENSE.txt"
   ],

package/dist/align/__tests__/align.test.cjs

@@ -1,283 +0,0 @@
-"use strict";
-var __create = Object.create;
-var __defProp = Object.defineProperty;
-var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
-var __getOwnPropNames = Object.getOwnPropertyNames;
-var __getProtoOf = Object.getPrototypeOf;
-var __hasOwnProp = Object.prototype.hasOwnProperty;
-var __knownSymbol = (name, symbol) => (symbol = Symbol[name]) ? symbol : Symbol.for("Symbol." + name);
-var __typeError = (msg) => {
-  throw TypeError(msg);
-};
-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 __toESM = (mod, isNodeMode, target) => (target = mod != null ? __create(__getProtoOf(mod)) : {}, __copyProps(
-  // If the importer is in node compatibility mode or this is not an ESM
-  // file that has been converted to a CommonJS file using a Babel-
-  // compatible transform (i.e. "__esModule" has not been set), then set
-  // "default" to the CommonJS "module.exports" for node compatibility.
-  isNodeMode || !mod || !mod.__esModule ? __defProp(target, "default", { value: mod, enumerable: true }) : target,
-  mod
-));
-var __using = (stack, value, async) => {
-  if (value != null) {
-    if (typeof value !== "object" && typeof value !== "function") __typeError("Object expected");
-    var dispose, inner;
-    if (async) dispose = value[__knownSymbol("asyncDispose")];
-    if (dispose === void 0) {
-      dispose = value[__knownSymbol("dispose")];
-      if (async) inner = dispose;
-    }
-    if (typeof dispose !== "function") __typeError("Object not disposable");
-    if (inner) dispose = function() {
-      try {
-        inner.call(this);
-      } catch (e) {
-        return Promise.reject(e);
-      }
-    };
-    stack.push([async, dispose, value]);
-  } else if (async) {
-    stack.push([async]);
-  }
-  return value;
-};
-var __callDispose = (stack, error, hasError) => {
-  var E = typeof SuppressedError === "function" ? SuppressedError : function(e, s, m, _) {
-    return _ = Error(m), _.name = "SuppressedError", _.error = e, _.suppressed = s, _;
-  };
-  var fail = (e) => error = hasError ? new E(e, error, "An error was suppressed during disposal") : (hasError = true, e);
-  var next = (it2) => {
-    while (it2 = stack.pop()) {
-      try {
-        var result = it2[1] && it2[1].call(it2[2]);
-        if (it2[0]) return Promise.resolve(result).then(next, (e) => (fail(e), next()));
-      } catch (e) {
-        fail(e);
-      }
-    }
-    if (hasError) throw error;
-  };
-  return next();
-};
-var import_node_assert = __toESM(require("node:assert"), 1);
-var import_promises = require("node:fs/promises");
-var import_node_path = require("node:path");
-var import_posix = require("node:path/posix");
-var import_node_test = require("node:test");
-var import_audiobook = require("@storyteller-platform/audiobook");
-var import_epub = require("@storyteller-platform/epub");
-var import_logging = require("../../common/logging.cjs");
-var import_segmentation = require("../../markup/segmentation.cjs");
-var import_align = require("../align.cjs");
-function createTestLogger() {
-  return (0, import_logging.createLogger)(process.env["CI"] ? "silent" : "info");
-}
-function sanitizeFilename(title) {
-  return title.replace(/[/\\:*?"<>|]/g, "-").replace(/\s+/g, " ").trim().replace(/[.]+$/, "");
-}
-function truncate(input, byteLimit, suffix = "") {
-  const normalized = input.normalize("NFC");
-  const encoder = new TextEncoder();
-  let result = "";
-  for (const char of normalized) {
-    const withSuffix = result + char + suffix;
-    const byteLength = encoder.encode(withSuffix).length;
-    if (byteLength > byteLimit) break;
-    result += char;
-  }
-  return result + suffix;
-}
-function getSafeFilepathSegment(name, suffix = "") {
-  return truncate(sanitizeFilename(name), 150, suffix);
-}
-async function assertAlignSnapshot(context, epub, transcriptionFilepaths) {
-  var _a, _b, _c, _d, _e, _f;
-  const snapshotFilename = getSafeFilepathSegment(context.fullName, ".snapshot");
-  const snapshotFilepath = (0, import_node_path.join)(
-    "src",
-    "align",
-    "__snapshots__",
-    snapshotFilename
-  );
-  let newSnapshot = "";
-  const manifest = await epub.getManifest();
-  const mediaOverlayItems = Object.values(manifest).map((item) => item.mediaOverlay).filter((mediaOverlayId) => !!mediaOverlayId).map((id) => manifest[id]);
-  const mediaOverlays = [];
-  for (const item of mediaOverlayItems) {
-    const contents = await epub.readItemContents(item.id, "utf-8");
-    const parsed = import_epub.Epub.xmlParser.parse(contents);
-    mediaOverlays.push(parsed);
-    const smil = import_epub.Epub.findXmlChildByName("smil", parsed);
-    if (!smil) continue;
-    const body = import_epub.Epub.findXmlChildByName("body", import_epub.Epub.getXmlChildren(smil));
-    if (!body) continue;
-    const seq = import_epub.Epub.findXmlChildByName("seq", import_epub.Epub.getXmlChildren(body));
-    if (!seq) continue;
-    const textref = (_a = seq[":@"]) == null ? void 0 : _a["@_epub:textref"];
-    if (!textref) continue;
-    newSnapshot += `// ${(0, import_posix.basename)(textref)}
-
-`;
-    const chapterContents = await epub.readFileContents(
-      textref,
-      item.href,
-      "utf-8"
-    );
-    const chapterXml = import_epub.Epub.xhtmlParser.parse(chapterContents);
-    const { result: segmentation } = await (0, import_segmentation.getXhtmlSegmentation)(
-      import_epub.Epub.getXhtmlBody(chapterXml),
-      {
-        primaryLocale: new Intl.Locale("en-US")
-      }
-    );
-    const chapterSentences = segmentation.map((s) => s.text).filter((s) => s.match(/\S/));
-    for (const par of import_epub.Epub.getXmlChildren(seq)) {
-      newSnapshot += `
-`;
-      const text = import_epub.Epub.findXmlChildByName("text", import_epub.Epub.getXmlChildren(par));
-      if (!text) continue;
-      const audio = import_epub.Epub.findXmlChildByName("audio", import_epub.Epub.getXmlChildren(par));
-      if (!audio) continue;
-      const textSrc = (_b = text[":@"]) == null ? void 0 : _b["@_src"];
-      if (!textSrc) continue;
-      const sentenceId = (_c = textSrc.match(/[0-9]+$/)) == null ? void 0 : _c[0];
-      if (sentenceId === void 0) continue;
-      const textSentence = chapterSentences[parseInt(sentenceId)];
-      if (!textSentence) continue;
-      newSnapshot += `Text:  ${textSentence.replace(/\n/, "")}
-`;
-      const audioSrc = (_d = audio[":@"]) == null ? void 0 : _d["@_src"];
-      if (!audioSrc) continue;
-      const audioStart = (_e = audio[":@"]) == null ? void 0 : _e["@_clipBegin"];
-      const audioEnd = (_f = audio[":@"]) == null ? void 0 : _f["@_clipEnd"];
-      if (!audioStart || !audioEnd) continue;
-      const audioStartTime = parseFloat(audioStart.slice(0, -1)) - 2e-3;
-      const audioEndTime = parseFloat(audioEnd.slice(0, -1));
-      const audioFilename = (0, import_posix.basename)(audioSrc, (0, import_node_path.extname)(audioSrc));
-      const transcriptionFilepath = transcriptionFilepaths.find(
-        (f) => (0, import_node_path.basename)(f, (0, import_node_path.extname)(f)) === audioFilename
-      );
-      if (!transcriptionFilepath) continue;
-      const transcription = JSON.parse(
-        await (0, import_promises.readFile)(transcriptionFilepath, { encoding: "utf-8" })
-      );
-      const transcriptionWords = [];
-      let started = false;
-      let i = 0;
-      let word = transcription.timeline[i];
-      while (word && word.endTime <= audioEndTime) {
-        if (word.startTime >= audioStartTime) {
-          started = true;
-        }
-        if (started) {
-          transcriptionWords.push(word.text);
-        }
-        word = transcription.timeline[++i];
-      }
-      const transcriptionSentence = transcriptionWords.join(" ");
-      newSnapshot += `Audio: ${transcriptionSentence}
-`;
-    }
-    newSnapshot += `
-`;
-  }
-  if (process.env["UPDATE_SNAPSHOTS"]) {
-    await (0, import_promises.mkdir)((0, import_node_path.dirname)(snapshotFilepath), { recursive: true });
-    await (0, import_promises.writeFile)(snapshotFilepath, newSnapshot, { encoding: "utf-8" });
-    return;
-  }
-  try {
-    const existingSnapshot = await (0, import_promises.readFile)(snapshotFilepath, {
-      encoding: "utf-8"
-    });
-    const existingLines = existingSnapshot.split("\n");
-    const newLines = newSnapshot.split("\n");
-    for (let i = 0; i < existingLines.length; i++) {
-      const existingLine = existingLines[i];
-      const newLine = newLines[i];
-      if (existingLine !== newLine) {
-        import_node_assert.default.strictEqual(
-          newLines.slice(Math.max(0, i - 5), i + 5),
-          existingLines.slice(Math.max(0, i - 5), i + 5)
-        );
-      }
-    }
-  } catch (e) {
-    if (e instanceof import_node_assert.default.AssertionError) {
-      throw e;
-    }
-    throw new import_node_assert.default.AssertionError({
-      actual: newSnapshot,
-      expected: "",
-      diff: "simple"
-    });
-  }
-}
-void (0, import_node_test.describe)("align", () => {
-  void (0, import_node_test.it)("should align Peter and Wendy", async (context) => {
-    var _stack = [];
-    try {
-      const epub = __using(_stack, await import_epub.Epub.from(
-        (0, import_node_path.join)(
-          "src",
-          "align",
-          "__fixtures__",
-          "peter-and-wendy",
-          "text",
-          "Peter and Wendy.epub"
-        )
-      ));
-      const audiobookDir = (0, import_node_path.join)(
-        "src",
-        "align",
-        "__fixtures__",
-        "peter-and-wendy",
-        "audio"
-      );
-      const audiobookFiles = await (0, import_promises.readdir)(audiobookDir).then(
-        (filenames) => filenames.filter((f) => (0, import_audiobook.isAudioFile)(f)).map((f) => (0, import_node_path.join)(audiobookDir, f))
-      );
-      const transcriptionsDir = (0, import_node_path.join)(
-        "src",
-        "align",
-        "__fixtures__",
-        "peter-and-wendy",
-        "transcriptions"
-      );
-      const transcriptionFilepaths = await (0, import_promises.readdir)(transcriptionsDir).then(
-        (filenames) => filenames.filter((f) => f.endsWith(".json")).map((f) => (0, import_node_path.join)(transcriptionsDir, f))
-      );
-      const transcriptions = await Promise.all(
-        transcriptionFilepaths.map(
-          async (p) => (0, import_promises.readFile)(p, { encoding: "utf-8" })
-        )
-      ).then(
-        (contents) => contents.map(
-          (c) => JSON.parse(c)
-        )
-      );
-      const aligner = new import_align.Aligner(
-        epub,
-        audiobookFiles,
-        transcriptions,
-        "sentence",
-        void 0,
-        createTestLogger()
-      );
-      const timing = await aligner.alignBook();
-      if (!process.env["CI"]) timing.print();
-      await assertAlignSnapshot(context, epub, transcriptionFilepaths);
-    } catch (_) {
-      var _error = _, _hasError = true;
-    } finally {
-      __callDispose(_stack, _error, _hasError);
-    }
-  });
-});

package/dist/align/__tests__/align.test.d.cts

@@ -1,2 +0,0 @@
-
-export {  }

package/dist/align/__tests__/align.test.d.ts

@@ -1,2 +0,0 @@
-
-export {  }