Fix align failed in win32 (#1205)

* upgrade deps

* upgrade yarn

* fix bugs caused by config

* prepare bump

* downgrade echogarden for test

* upgrade deps

* db backup may failed

* add more tests

.github/workflows/test-enjoy-app.yml

@@ -54,7 +54,7 @@ jobs:
           xvfb-run --auto-servernum --server-args="-screen 0 1280x960x24" -- yarn enjoy:test:main
 
       - name: Run main tests on Windows
-        # continue-on-error: true
+        continue-on-error: true
         if: contains(matrix.os, 'windows')
         run: yarn enjoy:test:main
 

.yarnrc.yml

@@ -4,4 +4,4 @@ nmHoistingLimits: workspaces
 
 nodeLinker: node-modules
 
-yarnPath: .yarn/releases/yarn-4.5.1.cjs
+yarnPath: .yarn/releases/yarn-4.5.2.cjs

1000h-portal/package.json

@@ -10,11 +10,11 @@
     "postinstall": "nuxt prepare"
   },
   "dependencies": {
-    "@nuxtjs/seo": "^2.0.0-rc.23",
-    "nuxt": "^3.14.159",
-    "nuxt-og-image": "^3.0.8",
+    "@nuxtjs/seo": "^2.0.2",
+    "nuxt": "^3.14.1592",
+    "nuxt-og-image": "^4.0.0",
     "vue": "^3.5.13",
-    "vue-router": "^4.4.5"
+    "vue-router": "^4.5.0"
   },
   "devDependencies": {
     "autoprefixer": "^10.4.20",

enjoy/e2e/main.spec.ts

@@ -90,6 +90,14 @@ test("validate echogarden recognition by whisper.cpp", async () => {
   expect(res.success).toBeTruthy();
 });
 
+test("validate echogarden alignment", async () => {
+  const res = await page.evaluate(() => {
+    return window.__ENJOY_APP__.echogarden.checkAlign();
+  });
+  console.info(res.log);
+  expect(res.success).toBeTruthy();
+});
+
 test("valid ffmpeg command", async () => {
   const res = await page.evaluate(() => {
     return window.__ENJOY_APP__.ffmpeg.check();

enjoy/package.json

@@ -3,7 +3,7 @@
   "private": true,
   "name": "enjoy",
   "productName": "Enjoy",
-  "version": "0.7.3",
+  "version": "0.7.4-preview",
   "description": "Enjoy desktop app",
   "main": ".vite/build/main.js",
   "types": "./src/types.d.ts",
@@ -53,7 +53,7 @@
     "@types/mark.js": "^8.11.12",
     "@types/mime-types": "^2.1.4",
     "@types/mustache": "^4.2.5",
-    "@types/node": "^22.9.0",
+    "@types/node": "^22.10.0",
     "@types/prop-types": "^15.7.13",
     "@types/rails__actioncable": "^6.1.11",
     "@types/react": "^18.3.12",
@@ -62,8 +62,8 @@
     "@types/unzipper": "^0.10.10",
     "@types/validator": "^13.12.2",
     "@types/wavesurfer.js": "^6.0.12",
-    "@typescript-eslint/eslint-plugin": "^8.15.0",
-    "@typescript-eslint/parser": "^8.15.0",
+    "@typescript-eslint/eslint-plugin": "^8.16.0",
+    "@typescript-eslint/parser": "^8.16.0",
     "@vitejs/plugin-react": "^4.3.3",
     "autoprefixer": "^10.4.20",
     "electron": "^33.2.0",
@@ -76,14 +76,14 @@
     "octokit": "^4.0.2",
     "progress": "^2.0.3",
     "prop-types": "^15.8.1",
-    "tailwind-merge": "^2.5.4",
+    "tailwind-merge": "^2.5.5",
     "tailwind-scrollbar": "^3.1.0",
     "tailwindcss": "^3.4.15",
     "tailwindcss-animate": "^1.0.7",
     "ts-node": "^10.9.2",
     "tslib": "^2.8.1",
-    "typescript": "^5.6.3",
-    "vite": "^5.4.11",
+    "typescript": "^5.7.2",
+    "vite": "^6.0.0",
     "vite-plugin-static-copy": "^2.1.0",
     "zx": "^8.2.2"
   },
@@ -92,8 +92,8 @@
     "@divisey/js-mdict": "^5.0.0",
     "@electron-forge/publisher-s3": "^7.5.0",
     "@hookform/resolvers": "^3.9.1",
-    "@langchain/community": "^0.3.15",
-    "@langchain/core": "^0.3.18",
+    "@langchain/community": "^0.3.16",
+    "@langchain/core": "^0.3.19",
     "@langchain/ollama": "^0.1.2",
     "@mozilla/readability": "^0.5.0",
     "@radix-ui/react-accordion": "^1.2.1",
@@ -126,12 +126,12 @@
     "@vidstack/react": "^1.12.12",
     "ahoy.js": "^0.4.4",
     "autosize": "^6.0.1",
-    "axios": "^1.7.7",
+    "axios": "^1.7.8",
     "camelcase": "^8.0.0",
     "camelcase-keys": "^9.1.3",
     "chart.js": "^4.4.6",
     "cheerio": "^1.0.0",
-    "class-variance-authority": "^0.7.0",
+    "class-variance-authority": "^0.7.1",
     "clsx": "^2.1.1",
     "cmdk": "^1.0.4",
     "command-exists": "^1.2.9",
@@ -141,9 +141,9 @@
     "dayjs": "^1.11.13",
     "decamelize": "^6.0.0",
     "decamelize-keys": "^2.0.1",
-    "echogarden": "^2.0.7",
+    "echogarden": "<2.0",
     "electron-context-menu": "^4.0.4",
-    "electron-log": "^5.2.2",
+    "electron-log": "^5.2.3",
     "electron-settings": "^4.0.4",
     "electron-squirrel-startup": "^1.0.1",
     "ffmpeg-static": "^5.2.0",
@@ -153,21 +153,21 @@
     "fs-extra": "^11.2.0",
     "html-to-text": "^9.0.5",
     "https-proxy-agent": "^7.0.5",
-    "i18next": "^23.16.5",
+    "i18next": "^24.0.2",
     "input-otp": "^1.4.1",
     "intl-tel-input": "^24.7.0",
     "js-md5": "^0.8.3",
     "langchain": "^0.3.6",
     "lodash": "^4.17.21",
     "lru-cache": "^11.0.2",
-    "lucide-react": "^0.460.0",
+    "lucide-react": "^0.461.0",
     "mark.js": "^8.11.1",
     "media-captions": "^0.0.18",
     "microsoft-cognitiveservices-speech-sdk": "^1.41.0",
     "mime-types": "^2.1.35",
     "mustache": "^4.2.0",
     "next-themes": "^0.4.3",
-    "openai": "^4.72.0",
+    "openai": "^4.73.1",
     "pitchfinder": "^2.3.2",
     "postcss": "^8.4.49",
     "proxy-agent": "^6.4.0",
@@ -179,10 +179,10 @@
     "react-frame-component": "^5.2.7",
     "react-hook-form": "^7.53.2",
     "react-hotkeys-hook": "^4.6.1",
-    "react-i18next": "^15.1.1",
+    "react-i18next": "^15.1.2",
     "react-markdown": "^9.0.1",
     "react-resizable-panels": "^2.1.7",
-    "react-router-dom": "^6.28.0",
+    "react-router-dom": "^7.0.1",
     "react-shadow-root": "^6.2.0",
     "react-tooltip": "^5.28.0",
     "reflect-metadata": "^0.2.2",
@@ -198,7 +198,7 @@
     "umzug": "^3.8.2",
     "unzipper": "^0.12.3",
     "update-electron-app": "^3.0.0",
-    "wavesurfer.js": "^7.8.8",
+    "wavesurfer.js": "^7.8.9",
     "zod": "^3.23.8",
     "zod-to-json-schema": "^3.23.5"
   }

enjoy/src/main/db/index.ts

@@ -175,7 +175,10 @@ db.connect = async () => {
     if (pendingMigrations.length > 0) {
       try {
         await db.backup({ force: true });
-
+      } catch (err) {
+        logger.error(err);
+      }
+      try {
         // migrate up to the latest state
         await umzug.up();
       } catch (err) {

enjoy/src/main/echogarden.ts

@@ -57,6 +57,9 @@ class EchogardenWrapper {
 
   constructor() {
     this.recognize = (sampleFile: string, options: RecognitionOptions) => {
+      if (!options) {
+        throw new Error("No config options provided");
+      }
       return new Promise((resolve, reject) => {
         const handler = (reason: any) => {
           // Remove the handler after it's triggered
@@ -89,6 +92,9 @@ class EchogardenWrapper {
       });
     };
     this.align = (input, transcript, options) => {
+      if (!options) {
+        throw new Error("No config options provided");
+      }
       return new Promise((resolve, reject) => {
         const handler = (reason: any) => {
           // Remove the handler after it's triggered
@@ -109,6 +115,9 @@ class EchogardenWrapper {
       });
     };
     this.alignSegments = (input, timeline, options) => {
+      if (!options) {
+        throw new Error("No config options provided");
+      }
       return new Promise((resolve, reject) => {
         const handler = (reason: any) => {
           // Remove the handler after it's triggered
@@ -139,8 +148,8 @@ class EchogardenWrapper {
       wordTimelineToSegmentSentenceTimeline;
   }
 
-  async check(
-    options: RecognitionOptions = {
+  async check(options: RecognitionOptions) {
+    options = options || {
       engine: "whisper",
       whisper: {
         model: "tiny.en",
@@ -148,8 +157,7 @@ class EchogardenWrapper {
       whisperCpp: {
         model: "tiny.en",
       },
-    }
-  ) {
+    };
     const sampleFile = path.join(__dirname, "samples", "jfk.wav");
 
     try {
@@ -174,12 +182,33 @@ class EchogardenWrapper {
     }
   }
 
+  async checkAlign(options: AlignmentOptions) {
+    options = options || {
+      language: "en",
+    };
+    const sampleFile = path.join(__dirname, "samples", "jfk.wav");
+    const transcript =
+      "And so my fellow Americans ask not what your country can do for you ask what you can do for your country.";
+    try {
+      const timeline = await this.align(sampleFile, transcript, options);
+      logger.info("timeline:", !!timeline);
+      return { success: true, log: "" };
+    } catch (e) {
+      logger.error(e);
+      return { success: false, log: e.message };
+    }
+  }
+
   /**
    * Transcodes the audio file at the enjoy:// protocol URL into a WAV format.
    * @param url - The URL of the audio file to transcode.
    * @returns A promise that resolves to the enjoy:// protocal URL of the transcoded WAV file.
    */
-  async transcode(url: string, sampleRate = 16000): Promise<string> {
+  async transcode(
+    url: string,
+    sampleRate: number | null = 16000
+  ): Promise<string> {
+    sampleRate = sampleRate || 16000;
     logger.info("echogarden-transcode:", url, sampleRate);
     const filePath = enjoyUrlToPath(url);
     const rawAudio = await this.ensureRawAudio(filePath, sampleRate);
@@ -295,6 +324,11 @@ class EchogardenWrapper {
       return this.check(options);
     });
 
+    ipcMain.handle("echogarden-check-align", async (_event, options: any) => {
+      logger.info("echogarden-check-align:", options);
+      return this.checkAlign(options);
+    });
+
     ipcMain.handle("echogarden-get-packages-dir", async (_event) => {
       return ensureAndGetPackagesDir();
     });

enjoy/src/preload.ts

@@ -3,7 +3,10 @@
 import { contextBridge, ipcRenderer, IpcRendererEvent } from "electron";
 import { version } from "../package.json";
 import { Timeline } from "echogarden/dist/utilities/Timeline";
-import { RecognitionOptions } from "echogarden/dist/api/API";
+import {
+  type AlignmentOptions,
+  type RecognitionOptions,
+} from "echogarden/dist/api/API";
 
 contextBridge.exposeInMainWorld("__ENJOY_APP__", {
   app: {
@@ -546,6 +549,9 @@ contextBridge.exposeInMainWorld("__ENJOY_APP__", {
     check: (options: RecognitionOptions) => {
       return ipcRenderer.invoke("echogarden-check", options);
     },
+    checkAlign: (options: AlignmentOptions) => {
+      return ipcRenderer.invoke("echogarden-check-align", options);
+    },
   },
   ffmpeg: {
     check: () => {

enjoy/src/renderer/context/ai-settings-provider.tsx

@@ -6,6 +6,9 @@ import {
 import { SttEngineOptionEnum, UserSettingKeyEnum } from "@/types/enums";
 import { GPT_PROVIDERS, TTS_PROVIDERS } from "@renderer/components";
 import { WHISPER_MODELS } from "@/constants";
+import log from "electron-log/renderer";
+
+const logger = log.scope("ai-settings-provider.tsx");
 
 type AISettingsProviderState = {
   sttEngine?: SttEngineOptionEnum;
@@ -118,9 +121,8 @@ export const AISettingsProvider = ({
 
     if (!config) {
       let model = "tiny";
-      const whisperModel = await EnjoyApp.userSettings.get(
-        UserSettingKeyEnum.WHISPER
-      );
+      const whisperModel =
+        (await EnjoyApp.userSettings.get(UserSettingKeyEnum.WHISPER)) || "";
       if (WHISPER_MODELS.includes(whisperModel)) {
         model = whisperModel;
       } else {

enjoy/src/renderer/hooks/use-transcribe.tsx

@@ -118,6 +118,7 @@ export const useTranscribe = () => {
         new Uint8Array(await blob.arrayBuffer()),
         segmentTimeline,
         {
+          engine: "dtw",
           language: language.split("-")[0],
           isolate,
         }
@@ -141,6 +142,7 @@ export const useTranscribe = () => {
         new Uint8Array(await blob.arrayBuffer()),
         transcript,
         {
+          engine: "dtw",
           language: language.split("-")[0],
           isolate,
         }
@@ -263,6 +265,7 @@ export const useTranscribe = () => {
       throw new Error(t("whisperTranscribeFailed", { error: err.message }));
     }
 
+    setOutput("Whisper transcribe done");
     const { transcript, timeline } = res;
 
     return {
@@ -306,7 +309,7 @@ export const useTranscribe = () => {
         timestamp_granularities: ["word", "segment"],
       })) as any;
 
-      setOutput("Aligning the transcript...");
+      setOutput("OpenAI transcribe done");
       const segmentTimeline = (res.segments || []).map((segment) => {
         return {
           type: "segment" as TimelineEntryType,
@@ -352,6 +355,7 @@ export const useTranscribe = () => {
         )
       ).data;
 
+      setOutput("Cloudflare transcribe done");
       const segmentTimeline: TimelineEntry[] = [];
       if (res.vtt) {
         const caption = await parseText(res.vtt, { type: "vtt" });
@@ -426,17 +430,17 @@ export const useTranscribe = () => {
 
         reco.canceled = (_s, e) => {
           if (e.reason === sdk.CancellationReason.Error) {
-            logger.error("CANCELED: Reason=" + e.reason);
+            logger.error("Azure transcribe canceled: Reason=" + e.reason);
             return reject(new Error(e.errorDetails));
           }
 
           reco.stopContinuousRecognitionAsync();
-          logger.info("CANCELED: Reason=" + e.reason);
+          logger.info("Azure transcribe canceled: Reason=" + e.reason);
         };
 
         reco.sessionStopped = async (_s, e) => {
           logger.info(
-            "Session stopped. Stop continuous recognition.",
+            "Azure transcribe session stopped. Stop continuous recognition.",
             e.sessionId
           );
           reco.stopContinuousRecognitionAsync();

enjoy/src/types/enjoy-app.d.ts

@@ -311,6 +311,10 @@ type EnjoyAppType = {
     ) => Promise<Timeline>;
     transcode: (input: string) => Promise<string>;
     check: (options?: any) => Promise<{ success: boolean; log: string }>;
+    checkAlign: (options?: any) => Promise<{
+      success: boolean;
+      log: string;
+    }>;
   };
   ffmpeg: {
     check: () => Promise<boolean>;

package.json

@@ -24,7 +24,7 @@
     "docs:preview": "yarn workspace 1000-hours preview",
     "portal:generate": "yarn workspace 1000h-portal generate"
   },
-  "packageManager": "yarn@4.5.1",
+  "packageManager": "yarn@4.5.2",
   "engines": {
     "node": ">=18.0.0"
   }