@fonoster/apiserver 0.7.36 → 0.7.37

package/dist/voice/tts/ElevenLabs.d.ts

@@ -21,6 +21,7 @@ declare class ElevenLabs extends AbstractTextToSpeech<typeof ENGINE_NAME> {
         ref: string;
         stream: Readable;
     }>;
+    private doSynthesize;
     static getConfigValidationSchema(): z.Schema;
     static getCredentialsValidationSchema(): z.Schema;
 }

package/dist/voice/tts/ElevenLabs.js

@@ -43,12 +43,32 @@ var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, ge
 };
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.ElevenLabs = exports.ENGINE_NAME = void 0;
-const common_1 = require("@fonoster/common");
+/*
+ * Copyright (C) 2024 by Fonoster Inc (https://fonoster.com)
+ * http://github.com/fonoster/fonoster
+ *
+ * This file is part of Fonoster
+ *
+ * Licensed under the MIT License (the "License");
+ * you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ *    https://opensource.org/licenses/MIT
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+const stream_1 = require("stream");
 const logger_1 = require("@fonoster/logger");
 const elevenlabs_1 = require("elevenlabs");
 const z = __importStar(require("zod"));
 const AbstractTextToSpeech_1 = require("./AbstractTextToSpeech");
 const isSsml_1 = require("./isSsml");
+const streamToBuffer_1 = require("./streamToBuffer");
+const textChunksByFirstNaturalPause_1 = require("../handlers/utils/textChunksByFirstNaturalPause"); // Assuming this is the chunking function
 const ENGINE_NAME = "tts.elevenlabs";
 exports.ENGINE_NAME = ENGINE_NAME;
 const logger = (0, logger_1.getLogger)({ service: "apiserver", filePath: __filename });
@@ -65,27 +85,52 @@ class ElevenLabs extends AbstractTextToSpeech_1.AbstractTextToSpeech {
         return __awaiter(this, void 0, void 0, function* () {
             logger.verbose(`synthesize [input: ${text}, isSsml=${(0, isSsml_1.isSsml)(text)} options: ${JSON.stringify(options)}]`);
             const { voice } = this.engineConfig.config;
-            const audioStream = yield this.client.generate({
+            const ref = this.createMediaReference();
+            const chunks = (0, textChunksByFirstNaturalPause_1.textChunksByFirstNaturalPause)(text);
+            const stream = new stream_1.Readable({ read() { } });
+            const results = new Array(chunks.length);
+            let nextIndexToPush = 0;
+            function observeQueue() {
+                if (nextIndexToPush < results.length &&
+                    results[nextIndexToPush] !== undefined) {
+                    stream.push(results[nextIndexToPush]);
+                    nextIndexToPush++;
+                    setImmediate(observeQueue);
+                }
+                else if (nextIndexToPush < results.length) {
+                    setTimeout(observeQueue, 10);
+                }
+                else {
+                    stream.push(null);
+                }
+            }
+            observeQueue();
+            chunks.forEach((textChunk, index) => {
+                this.doSynthesize(textChunk, voice)
+                    .then((synthesizedText) => {
+                    results[index] = synthesizedText;
+                })
+                    .catch((error) => {
+                    stream.emit("error", error);
+                });
+            });
+            return { ref, stream };
+        });
+    }
+    doSynthesize(text, voice) {
+        return __awaiter(this, void 0, void 0, function* () {
+            const response = yield this.client.generate({
                 stream: true,
                 voice,
                 text,
-                // TODO: This should be configurable
                 model_id: "eleven_turbo_v2_5",
                 output_format: "pcm_16000"
             });
-            const ref = this.createMediaReference();
-            audioStream.on("error", (error) => {
-                logger.error(`Error reading file: ${error.message}`);
-            });
-            return { ref, stream: audioStream };
+            return (yield (0, streamToBuffer_1.streamToBuffer)(response));
         });
     }
     static getConfigValidationSchema() {
-        return z.object({
-            voice: z.nativeEnum(common_1.ElevenLabsVoice, {
-                message: "Invalid ElevenLabs voice."
-            })
-        });
+        return z.object({});
     }
     static getCredentialsValidationSchema() {
         return z.object({

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@fonoster/apiserver",
-  "version": "0.7.36",
+  "version": "0.7.37",
   "description": "APIServer for Fonoster",
   "author": "Pedro Sanders <psanders@fonoster.com>",
   "homepage": "https://github.com/fonoster/fonoster#readme",
@@ -21,12 +21,12 @@
   },
   "dependencies": {
     "@deepgram/sdk": "^3.5.1",
-    "@fonoster/common": "^0.7.36",
-    "@fonoster/identity": "^0.7.36",
-    "@fonoster/logger": "^0.7.36",
-    "@fonoster/sipnet": "^0.7.36",
-    "@fonoster/streams": "^0.7.36",
-    "@fonoster/types": "^0.7.36",
+    "@fonoster/common": "^0.7.37",
+    "@fonoster/identity": "^0.7.37",
+    "@fonoster/logger": "^0.7.37",
+    "@fonoster/sipnet": "^0.7.37",
+    "@fonoster/streams": "^0.7.37",
+    "@fonoster/types": "^0.7.37",
     "@google-cloud/speech": "^6.6.0",
     "@google-cloud/text-to-speech": "^5.3.0",
     "@grpc/grpc-js": "~1.10.6",
@@ -72,5 +72,5 @@
     "@types/uuid": "^9.0.8",
     "@types/validator": "^13.12.0"
   },
-  "gitHead": "616656596fe60b32caccad2e63f628fb35b139d0"
+  "gitHead": "b9696f512b3abfd6a798045b16ed9f102d58b11a"
 }