feature: 'kokoro-js' supports TTS #3412 (#3656)

* feature: 'kokoro-js' supports TTS #3412 * Linting, add credits for kokoro library * Fix voice preview * Fix display languages on previews * Fix settings restoration. Debounce model init on settings change * Fix engine sorting * Move TTS processing to a web worker. Remove unused gain setting * Speaking rate fix * Update status when recreating a worker * Pass voices list from TTS engine * Call dispose function on provider change * Extend worker init timeout to 10 minutes --------- Co-authored-by: ryan <1014670860@qq.com> Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
2025-06-05 21:59:27 +02:00 · 2025-03-11 04:54:54 +08:00
parent 13a3f4772e
commit e23f3a6314
11 changed files with 491 additions and 27 deletions
--- a/public/scripts/extensions/tts/kokoro-worker.js
+++ b/public/scripts/extensions/tts/kokoro-worker.js
@@ -0,0 +1,113 @@
+// kokoro-worker.js
+/** @type {import('./lib/kokoro.web.js').KokoroTTS} */
+let tts = null;
+/** @type {boolean} */
+let ready = false;
+/** @type {string[]} */
+let voices = [];
+
+// Handle messages from the main thread
+self.onmessage = async function(e) {
+    const { action, data } = e.data;
+
+    switch (action) {
+        case 'initialize':
+            try {
+                const result = await initializeTts(data);
+                self.postMessage({
+                    action: 'initialized',
+                    success: result,
+                    voices,
+                });
+            } catch (error) {
+                self.postMessage({
+                    action: 'initialized',
+                    success: false,
+                    error: error.message,
+                });
+            }
+            break;
+
+        case 'generateTts':
+            try {
+                const audioBlob = await generateTts(data.text, data.voice, data.speakingRate);
+                const blobUrl = URL.createObjectURL(audioBlob);
+                self.postMessage({
+                    action: 'generatedTts',
+                    success: true,
+                    blobUrl,
+                    requestId: data.requestId,
+                });
+            } catch (error) {
+                self.postMessage({
+                    action: 'generatedTts',
+                    success: false,
+                    error: error.message,
+                    requestId: data.requestId,
+                });
+            }
+            break;
+
+        case 'checkReady':
+            self.postMessage({ action: 'readyStatus', ready });
+            break;
+    }
+};
+
+// Initialize the TTS engine
+async function initializeTts(settings) {
+    try {
+        const { KokoroTTS } = await import('./lib/kokoro.web.js');
+
+        console.log('Worker: Initializing Kokoro TTS with settings:', {
+            modelId: settings.modelId,
+            dtype: settings.dtype,
+            device: settings.device,
+        });
+
+        // Create TTS instance
+        tts = await KokoroTTS.from_pretrained(settings.modelId, {
+            dtype: settings.dtype,
+            device: settings.device,
+        });
+
+        // Get available voices
+        voices = Object.keys(tts.voices);
+
+        // Check if generate method exists
+        if (typeof tts.generate !== 'function') {
+            throw new Error('TTS instance does not have generate method');
+        }
+
+        console.log('Worker: TTS initialized successfully');
+        ready = true;
+        return true;
+    } catch (error) {
+        console.error('Worker: Kokoro TTS initialization failed:', error);
+        ready = false;
+        throw error;
+    }
+}
+
+// Generate TTS audio
+async function generateTts(text, voiceId, speakingRate) {
+    if (!ready || !tts) {
+        throw new Error('TTS engine not initialized');
+    }
+
+    if (text.trim().length === 0) {
+        throw new Error('Empty text');
+    }
+
+    try {
+        const audio = await tts.generate(text, {
+            voice: voiceId,
+            speed: speakingRate || 1.0,
+        });
+
+        return audio.toBlob();
+    } catch (error) {
+        console.error('Worker: TTS generation failed:', error);
+        throw error;
+    }
+}