@mastra/voice-elevenlabs 0.10.6 → 0.10.7

package/dist/index.cjs

@@ -155,3 +155,5 @@ var ElevenLabsVoice = class extends voice.MastraVoice {
 };
 
 exports.ElevenLabsVoice = ElevenLabsVoice;
+//# sourceMappingURL=index.cjs.map
+//# sourceMappingURL=index.cjs.map

package/dist/index.cjs.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/index.ts"],"names":["MastraVoice","ElevenLabsClient","buffer","File"],"mappings":";;;;;;;AAmCO,IAAM,eAAA,GAAN,cAA8BA,iBAAA,CAAY;AAAA,EACvC,MAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWR,WAAA,CAAY;AAAA,IACV,WAAA;AAAA,IACA,cAAA;AAAA,IACA;AAAA,GACF,GAAuG,EAAC,EAAG;AACzG,IAAA,MAAM,MAAA,GAAS,WAAA,EAAa,MAAA,IAAU,OAAA,CAAQ,GAAA,CAAI,kBAAA;AAClD,IAAA,KAAA,CAAM;AAAA,MACJ,WAAA,EAAa;AAAA,QACX,IAAA,EAAM,aAAa,IAAA,IAAQ,wBAAA;AAAA,QAC3B,QAAQ,WAAA,EAAa;AAAA,OACvB;AAAA,MACA,cAAA,EAAgB;AAAA,QACd,IAAA,EAAM,gBAAgB,IAAA,IAAQ,WAAA;AAAA,QAC9B,QAAQ,cAAA,EAAgB;AAAA,OAC1B;AAAA,MACA;AAAA,KACD,CAAA;AAED,IAAA,IAAI,CAAC,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,MAAM,+BAA+B,CAAA;AAAA,IACjD;AAEA,IAAA,IAAA,CAAK,MAAA,GAAS,IAAIC,2BAAA,CAAiB;AAAA,MACjC;AAAA,KACD,CAAA;AAED,IAAA,IAAA,CAAK,UAAU,OAAA,IAAW,sBAAA;AAAA,EAC5B;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EASA,MAAM,WAAA,GAAc;AAClB,IAAA,MAAM,GAAA,GAAM,MAAM,IAAA,CAAK,MAAA,CAAO,YAAY;AACxC,MAAA,MAAM,MAAA,GAAS,MAAM,IAAA,CAAK,MAAA,CAAO,OAAO,MAAA,EAAO;AAC/C,MAAA,OACE,MAAA,EAAQ,MAAA,EAAQ,GAAA,CAAI,CAAA,KAAA,MAAU;AAAA,QAC5B,SAAS,KAAA,CAAM,QAAA;AAAA,QACf,MAAM,KAAA,CAAM,IAAA;AAAA,QACZ,QAAA,EAAU,KAAA,CAAM,MAAA,EAAQ,QAAA,IAAY,IAAA;AAAA,QACpC,MAAA,EAAQ,KAAA,CAAM,MAAA,EAAQ,MAAA,IAAU;AAAA,OAClC,CAAE,KAAK,EAAC;AAAA,IAEZ,CAAA,EAAG,yBAAyB,CAAA,EAAE;AAE9B,IAAA,OAAO,GAAA;AAAA,EACT;AAAA,EAEA,MAAc,eAAe,MAAA,EAAgD;AAC3E,IAAA,MAAM,SAAmB,EAAC;AAC1B,IAAA,WAAA,MAAiB,SAAS,MAAA,EAAQ;AAChC,MAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,QAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,MAChC,CAAA,MAAO;AACL,QAAA,MAAA,CAAO,KAAK,KAAK,CAAA;AAAA,MACnB;AAAA,IACF;AACA,IAAA,OAAO,MAAA,CAAO,MAAA,CAAO,MAAM,CAAA,CAAE,SAAS,OAAO,CAAA;AAAA,EAC/C;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAaA,MAAM,KAAA,CAAM,KAAA,EAAuC,OAAA,EAAgE;AACjH,IAAA,MAAM,OAAA,GAAU,OAAA,EAAS,OAAA,IAAW,IAAA,CAAK,OAAA;AACzC,IAAA,IAAI,CAAC,OAAA,EAAS;AACZ,MAAA,MAAM,IAAI,MAAM,sBAAsB,CAAA;AAAA,IACxC;AAEA,IAAA,IAAI,CAAC,IAAA,CAAK,WAAA,EAAa,IAAA,EAAM;AAC3B,MAAA,MAAM,IAAI,MAAM,2BAA2B,CAAA;AAAA,IAC7C;AACA,IAAA,MAAM,IAAA,GAAO,OAAO,KAAA,KAAU,QAAA,GAAW,QAAQ,MAAM,IAAA,CAAK,eAAe,KAAK,CAAA;AAChF,IAAA,MAAM,GAAA,GAAM,MAAM,IAAA,CAAK,MAAA,CAAO,YAAY;AACxC,MAAA,OAAO,MAAM,IAAA,CAAK,MAAA,CAAO,QAAA,CAAS;AAAA,QAChC,IAAA;AAAA,QACA,KAAA,EAAO,OAAA;AAAA,QACP,QAAA,EAAU,KAAK,WAAA,EAAa,IAAA;AAAA,QAC5B,MAAA,EAAQ;AAAA,OACT,CAAA;AAAA,IACH,CAAA,EAAG,wBAAwB,CAAA,EAAE;AAE7B,IAAA,OAAO,GAAA;AAAA,EACT;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAOA,MAAM,WAAA,GAAc;AAClB,IAAA,OAAO,EAAE,SAAS,IAAA,EAAK;AAAA,EACzB;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAkBA,MAAM,MAAA,CAAO,KAAA,EAA8B,OAAA,EAAoD;AAC7F,IAAA,MAAM,GAAA,GAAM,MAAM,IAAA,CAAK,MAAA,CAAO,YAAY;AACxC,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,WAAA,MAAiB,SAAS,KAAA,EAAO;AAC/B,QAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,UAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,QAChC,CAAA,MAAO;AACL,UAAA,MAAA,CAAO,KAAK,KAAK,CAAA;AAAA,QACnB;AAAA,MACF;AACA,MAAA,MAAMC,QAAA,GAAS,MAAA,CAAO,MAAA,CAAO,MAAM,CAAA;AAEnC,MAAA,MAAM,EAAE,eAAe,gBAAA,EAAkB,YAAA,EAAc,UAAU,GAAG,cAAA,EAAe,GAAI,OAAA,IAAW,EAAC;AAEnG,MAAA,MAAM,IAAA,GAAO,IAAIC,WAAA,CAAK,CAACD,QAAM,CAAA,EAAG,CAAA,MAAA,EAAS,QAAA,IAAY,KAAK,CAAA,CAAE,CAAA;AAE5D,MAAA,MAAM,aAAA,GAAgB,MAAM,IAAA,CAAK,MAAA,CAAO,YAAA,CAAa,OAAA;AAAA,QACnD;AAAA,UACE,IAAA;AAAA,UACA,QAAA,EAAU,KAAK,cAAA,EAAgB,IAAA;AAAA,UAC/B,aAAA;AAAA,UACA,gBAAA;AAAA,UACA;AAAA,SACF;AAAA,QACA;AAAA,OACF;AAEA,MAAA,OAAO,aAAA,CAAc,IAAA;AAAA,IACvB,CAAA,EAAG,yBAAyB,CAAA,EAAE;AAE9B,IAAA,OAAO,GAAA;AAAA,EACT;AACF","file":"index.cjs","sourcesContent":["import { File } from 'node:buffer';\nimport { MastraVoice } from '@mastra/core/voice';\nimport { ElevenLabsClient } from 'elevenlabs';\n\ntype ElevenLabsModel =\n  | 'eleven_multilingual_v2'\n  | 'eleven_flash_v2_5'\n  | 'eleven_flash_v2'\n  | 'eleven_multilingual_sts_v2'\n  | 'eleven_english_sts_v2'\n  | 'scribe_v1';\n\ninterface ElevenLabsVoiceConfig {\n  name?: ElevenLabsModel;\n  apiKey?: string;\n}\n\ninterface SpeechToTextOptions {\n  language_code?: string;\n  tag_audio_events?: boolean;\n  num_speakers?: number;\n  filetype?: string;\n}\n\ninterface RequestOptions {\n  timeoutInSeconds?: number;\n  maxRetries?: number;\n  abortSignal?: AbortSignal;\n  apiKey?: string | undefined;\n  headers?: Record<string, string>;\n}\n\n// Combined options type\ntype ElevenLabsListenOptions = SpeechToTextOptions & RequestOptions;\n\nexport class ElevenLabsVoice extends MastraVoice {\n  private client: ElevenLabsClient;\n\n  /**\n   * Creates an instance of the ElevenLabsVoice class.\n   *\n   * @param {Object} options - The options for the voice configuration.\n   * @param {ElevenLabsVoiceConfig} [options.speechModel] - The configuration for the speech model, including the model name and API key.\n   * @param {string} [options.speaker] - The ID of the speaker to use. If not provided, a default speaker will be used.\n   *\n   * @throws {Error} If the ELEVENLABS_API_KEY is not set in the environment variables.\n   */\n  constructor({\n    speechModel,\n    listeningModel,\n    speaker,\n  }: { speechModel?: ElevenLabsVoiceConfig; listeningModel?: ElevenLabsVoiceConfig; speaker?: string } = {}) {\n    const apiKey = speechModel?.apiKey ?? process.env.ELEVENLABS_API_KEY;\n    super({\n      speechModel: {\n        name: speechModel?.name ?? 'eleven_multilingual_v2',\n        apiKey: speechModel?.apiKey,\n      },\n      listeningModel: {\n        name: listeningModel?.name ?? 'scribe_v1',\n        apiKey: listeningModel?.apiKey,\n      },\n      speaker,\n    });\n\n    if (!apiKey) {\n      throw new Error('ELEVENLABS_API_KEY is not set');\n    }\n\n    this.client = new ElevenLabsClient({\n      apiKey,\n    });\n\n    this.speaker = speaker || '9BWtsMINqrJLrRacOk9x'; // Aria is the default speaker\n  }\n\n  /**\n   * Retrieves a list of available speakers from the Eleven Labs API.\n   * Each speaker includes their ID, name, language, and gender.\n   *\n   * @returns {Promise<Array<{ voiceId: string, name: string, language: string, gender: string }>>}\n   * A promise that resolves to an array of speaker objects.\n   */\n  async getSpeakers() {\n    const res = await this.traced(async () => {\n      const voices = await this.client.voices.getAll();\n      return (\n        voices?.voices?.map(voice => ({\n          voiceId: voice.voice_id,\n          name: voice.name,\n          language: voice.labels?.language || 'en',\n          gender: voice.labels?.gender || 'neutral',\n        })) ?? []\n      );\n    }, 'voice.elevenlabs.voices')();\n\n    return res;\n  }\n\n  private async streamToString(stream: NodeJS.ReadableStream): Promise<string> {\n    const chunks: Buffer[] = [];\n    for await (const chunk of stream) {\n      if (typeof chunk === 'string') {\n        chunks.push(Buffer.from(chunk));\n      } else {\n        chunks.push(chunk);\n      }\n    }\n    return Buffer.concat(chunks).toString('utf-8');\n  }\n\n  /**\n   * Converts text or audio input into speech using the Eleven Labs API.\n   *\n   * @param {string | NodeJS.ReadableStream} input - The text to be converted to speech or a stream containing audio data.\n   * @param {Object} [options] - Optional parameters for the speech generation.\n   * @param {string} [options.speaker] - The ID of the speaker to use for the speech. If not provided, the default speaker will be used.\n   *\n   * @returns {Promise<NodeJS.ReadableStream>} A promise that resolves to a readable stream of the generated speech.\n   *\n   * @throws {Error} If no speaker is specified or if no speech model is set.\n   */\n  async speak(input: string | NodeJS.ReadableStream, options?: { speaker?: string }): Promise<NodeJS.ReadableStream> {\n    const speaker = options?.speaker || this.speaker;\n    if (!speaker) {\n      throw new Error('No speaker specified');\n    }\n\n    if (!this.speechModel?.name) {\n      throw new Error('No speech model specified');\n    }\n    const text = typeof input === 'string' ? input : await this.streamToString(input);\n    const res = await this.traced(async () => {\n      return await this.client.generate({\n        text,\n        voice: speaker,\n        model_id: this.speechModel?.name as ElevenLabsModel,\n        stream: true,\n      });\n    }, 'voice.elevenlabs.speak')();\n\n    return res;\n  }\n\n  /**\n   * Checks if listening capabilities are enabled.\n   *\n   * @returns {Promise<{ enabled: boolean }>}\n   */\n  async getListener() {\n    return { enabled: true };\n  }\n\n  /**\n   * Converts audio input to text using ElevenLabs Speech-to-Text API.\n   *\n   * @param input - A readable stream containing the audio data to transcribe\n   * @param options - Configuration options for the transcription\n   * @param options.language_code - ISO language code (e.g., 'en', 'fr', 'es')\n   * @param options.tag_audio_events - Whether to tag audio events like [MUSIC], [LAUGHTER], etc.\n   * @param options.num_speakers - Number of speakers to detect in the audio\n   * @param options.filetype - Audio file format (e.g., 'mp3', 'wav', 'ogg')\n   * @param options.timeoutInSeconds - Request timeout in seconds\n   * @param options.maxRetries - Maximum number of retry attempts\n   * @param options.abortSignal - Signal to abort the request\n   *\n   * @returns A Promise that resolves to the transcribed text\n   *\n   */\n  async listen(input: NodeJS.ReadableStream, options?: ElevenLabsListenOptions): Promise<string> {\n    const res = await this.traced(async () => {\n      const chunks: Buffer[] = [];\n      for await (const chunk of input) {\n        if (typeof chunk === 'string') {\n          chunks.push(Buffer.from(chunk));\n        } else {\n          chunks.push(chunk);\n        }\n      }\n      const buffer = Buffer.concat(chunks);\n\n      const { language_code, tag_audio_events, num_speakers, filetype, ...requestOptions } = options || {};\n\n      const file = new File([buffer], `audio.${filetype || 'mp3'}`);\n\n      const transcription = await this.client.speechToText.convert(\n        {\n          file: file,\n          model_id: this.listeningModel?.name as ElevenLabsModel,\n          language_code,\n          tag_audio_events,\n          num_speakers,\n        },\n        requestOptions,\n      );\n\n      return transcription.text;\n    }, 'voice.elevenlabs.listen')();\n\n    return res;\n  }\n}\n"]}

package/dist/index.d.ts

@@ -1 +1,92 @@
-export { ElevenLabsVoice } from './_tsup-dts-rollup.js';
+import { MastraVoice } from '@mastra/core/voice';
+type ElevenLabsModel = 'eleven_multilingual_v2' | 'eleven_flash_v2_5' | 'eleven_flash_v2' | 'eleven_multilingual_sts_v2' | 'eleven_english_sts_v2' | 'scribe_v1';
+interface ElevenLabsVoiceConfig {
+    name?: ElevenLabsModel;
+    apiKey?: string;
+}
+interface SpeechToTextOptions {
+    language_code?: string;
+    tag_audio_events?: boolean;
+    num_speakers?: number;
+    filetype?: string;
+}
+interface RequestOptions {
+    timeoutInSeconds?: number;
+    maxRetries?: number;
+    abortSignal?: AbortSignal;
+    apiKey?: string | undefined;
+    headers?: Record<string, string>;
+}
+type ElevenLabsListenOptions = SpeechToTextOptions & RequestOptions;
+export declare class ElevenLabsVoice extends MastraVoice {
+    private client;
+    /**
+     * Creates an instance of the ElevenLabsVoice class.
+     *
+     * @param {Object} options - The options for the voice configuration.
+     * @param {ElevenLabsVoiceConfig} [options.speechModel] - The configuration for the speech model, including the model name and API key.
+     * @param {string} [options.speaker] - The ID of the speaker to use. If not provided, a default speaker will be used.
+     *
+     * @throws {Error} If the ELEVENLABS_API_KEY is not set in the environment variables.
+     */
+    constructor({ speechModel, listeningModel, speaker, }?: {
+        speechModel?: ElevenLabsVoiceConfig;
+        listeningModel?: ElevenLabsVoiceConfig;
+        speaker?: string;
+    });
+    /**
+     * Retrieves a list of available speakers from the Eleven Labs API.
+     * Each speaker includes their ID, name, language, and gender.
+     *
+     * @returns {Promise<Array<{ voiceId: string, name: string, language: string, gender: string }>>}
+     * A promise that resolves to an array of speaker objects.
+     */
+    getSpeakers(): Promise<{
+        voiceId: string;
+        name: string | undefined;
+        language: string;
+        gender: string;
+    }[]>;
+    private streamToString;
+    /**
+     * Converts text or audio input into speech using the Eleven Labs API.
+     *
+     * @param {string | NodeJS.ReadableStream} input - The text to be converted to speech or a stream containing audio data.
+     * @param {Object} [options] - Optional parameters for the speech generation.
+     * @param {string} [options.speaker] - The ID of the speaker to use for the speech. If not provided, the default speaker will be used.
+     *
+     * @returns {Promise<NodeJS.ReadableStream>} A promise that resolves to a readable stream of the generated speech.
+     *
+     * @throws {Error} If no speaker is specified or if no speech model is set.
+     */
+    speak(input: string | NodeJS.ReadableStream, options?: {
+        speaker?: string;
+    }): Promise<NodeJS.ReadableStream>;
+    /**
+     * Checks if listening capabilities are enabled.
+     *
+     * @returns {Promise<{ enabled: boolean }>}
+     */
+    getListener(): Promise<{
+        enabled: boolean;
+    }>;
+    /**
+     * Converts audio input to text using ElevenLabs Speech-to-Text API.
+     *
+     * @param input - A readable stream containing the audio data to transcribe
+     * @param options - Configuration options for the transcription
+     * @param options.language_code - ISO language code (e.g., 'en', 'fr', 'es')
+     * @param options.tag_audio_events - Whether to tag audio events like [MUSIC], [LAUGHTER], etc.
+     * @param options.num_speakers - Number of speakers to detect in the audio
+     * @param options.filetype - Audio file format (e.g., 'mp3', 'wav', 'ogg')
+     * @param options.timeoutInSeconds - Request timeout in seconds
+     * @param options.maxRetries - Maximum number of retry attempts
+     * @param options.abortSignal - Signal to abort the request
+     *
+     * @returns A Promise that resolves to the transcribed text
+     *
+     */
+    listen(input: NodeJS.ReadableStream, options?: ElevenLabsListenOptions): Promise<string>;
+}
+export {};
+//# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,WAAW,EAAE,MAAM,oBAAoB,CAAC;AAGjD,KAAK,eAAe,GAChB,wBAAwB,GACxB,mBAAmB,GACnB,iBAAiB,GACjB,4BAA4B,GAC5B,uBAAuB,GACvB,WAAW,CAAC;AAEhB,UAAU,qBAAqB;IAC7B,IAAI,CAAC,EAAE,eAAe,CAAC;IACvB,MAAM,CAAC,EAAE,MAAM,CAAC;CACjB;AAED,UAAU,mBAAmB;IAC3B,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,gBAAgB,CAAC,EAAE,OAAO,CAAC;IAC3B,YAAY,CAAC,EAAE,MAAM,CAAC;IACtB,QAAQ,CAAC,EAAE,MAAM,CAAC;CACnB;AAED,UAAU,cAAc;IACtB,gBAAgB,CAAC,EAAE,MAAM,CAAC;IAC1B,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,WAAW,CAAC,EAAE,WAAW,CAAC;IAC1B,MAAM,CAAC,EAAE,MAAM,GAAG,SAAS,CAAC;IAC5B,OAAO,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;CAClC;AAGD,KAAK,uBAAuB,GAAG,mBAAmB,GAAG,cAAc,CAAC;AAEpE,qBAAa,eAAgB,SAAQ,WAAW;IAC9C,OAAO,CAAC,MAAM,CAAmB;IAEjC;;;;;;;;OAQG;gBACS,EACV,WAAW,EACX,cAAc,EACd,OAAO,GACR,GAAE;QAAE,WAAW,CAAC,EAAE,qBAAqB,CAAC;QAAC,cAAc,CAAC,EAAE,qBAAqB,CAAC;QAAC,OAAO,CAAC,EAAE,MAAM,CAAA;KAAO;IAyBzG;;;;;;OAMG;IACG,WAAW;;;;;;YAgBH,cAAc;IAY5B;;;;;;;;;;OAUG;IACG,KAAK,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAAC,cAAc,EAAE,OAAO,CAAC,EAAE;QAAE,OAAO,CAAC,EAAE,MAAM,CAAA;KAAE,GAAG,OAAO,CAAC,MAAM,CAAC,cAAc,CAAC;IAsBlH;;;;OAIG;IACG,WAAW;;;IAIjB;;;;;;;;;;;;;;;OAeG;IACG,MAAM,CAAC,KAAK,EAAE,MAAM,CAAC,cAAc,EAAE,OAAO,CAAC,EAAE,uBAAuB,GAAG,OAAO,CAAC,MAAM,CAAC;CAgC/F"}

package/dist/index.js

@@ -153,3 +153,5 @@ var ElevenLabsVoice = class extends MastraVoice {
 };
 
 export { ElevenLabsVoice };
+//# sourceMappingURL=index.js.map
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/index.ts"],"names":[],"mappings":";;;;;AAmCO,IAAM,eAAA,GAAN,cAA8B,WAAA,CAAY;AAAA,EACvC,MAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWR,WAAA,CAAY;AAAA,IACV,WAAA;AAAA,IACA,cAAA;AAAA,IACA;AAAA,GACF,GAAuG,EAAC,EAAG;AACzG,IAAA,MAAM,MAAA,GAAS,WAAA,EAAa,MAAA,IAAU,OAAA,CAAQ,GAAA,CAAI,kBAAA;AAClD,IAAA,KAAA,CAAM;AAAA,MACJ,WAAA,EAAa;AAAA,QACX,IAAA,EAAM,aAAa,IAAA,IAAQ,wBAAA;AAAA,QAC3B,QAAQ,WAAA,EAAa;AAAA,OACvB;AAAA,MACA,cAAA,EAAgB;AAAA,QACd,IAAA,EAAM,gBAAgB,IAAA,IAAQ,WAAA;AAAA,QAC9B,QAAQ,cAAA,EAAgB;AAAA,OAC1B;AAAA,MACA;AAAA,KACD,CAAA;AAED,IAAA,IAAI,CAAC,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,MAAM,+BAA+B,CAAA;AAAA,IACjD;AAEA,IAAA,IAAA,CAAK,MAAA,GAAS,IAAI,gBAAA,CAAiB;AAAA,MACjC;AAAA,KACD,CAAA;AAED,IAAA,IAAA,CAAK,UAAU,OAAA,IAAW,sBAAA;AAAA,EAC5B;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EASA,MAAM,WAAA,GAAc;AAClB,IAAA,MAAM,GAAA,GAAM,MAAM,IAAA,CAAK,MAAA,CAAO,YAAY;AACxC,MAAA,MAAM,MAAA,GAAS,MAAM,IAAA,CAAK,MAAA,CAAO,OAAO,MAAA,EAAO;AAC/C,MAAA,OACE,MAAA,EAAQ,MAAA,EAAQ,GAAA,CAAI,CAAA,KAAA,MAAU;AAAA,QAC5B,SAAS,KAAA,CAAM,QAAA;AAAA,QACf,MAAM,KAAA,CAAM,IAAA;AAAA,QACZ,QAAA,EAAU,KAAA,CAAM,MAAA,EAAQ,QAAA,IAAY,IAAA;AAAA,QACpC,MAAA,EAAQ,KAAA,CAAM,MAAA,EAAQ,MAAA,IAAU;AAAA,OAClC,CAAE,KAAK,EAAC;AAAA,IAEZ,CAAA,EAAG,yBAAyB,CAAA,EAAE;AAE9B,IAAA,OAAO,GAAA;AAAA,EACT;AAAA,EAEA,MAAc,eAAe,MAAA,EAAgD;AAC3E,IAAA,MAAM,SAAmB,EAAC;AAC1B,IAAA,WAAA,MAAiB,SAAS,MAAA,EAAQ;AAChC,MAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,QAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,MAChC,CAAA,MAAO;AACL,QAAA,MAAA,CAAO,KAAK,KAAK,CAAA;AAAA,MACnB;AAAA,IACF;AACA,IAAA,OAAO,MAAA,CAAO,MAAA,CAAO,MAAM,CAAA,CAAE,SAAS,OAAO,CAAA;AAAA,EAC/C;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAaA,MAAM,KAAA,CAAM,KAAA,EAAuC,OAAA,EAAgE;AACjH,IAAA,MAAM,OAAA,GAAU,OAAA,EAAS,OAAA,IAAW,IAAA,CAAK,OAAA;AACzC,IAAA,IAAI,CAAC,OAAA,EAAS;AACZ,MAAA,MAAM,IAAI,MAAM,sBAAsB,CAAA;AAAA,IACxC;AAEA,IAAA,IAAI,CAAC,IAAA,CAAK,WAAA,EAAa,IAAA,EAAM;AAC3B,MAAA,MAAM,IAAI,MAAM,2BAA2B,CAAA;AAAA,IAC7C;AACA,IAAA,MAAM,IAAA,GAAO,OAAO,KAAA,KAAU,QAAA,GAAW,QAAQ,MAAM,IAAA,CAAK,eAAe,KAAK,CAAA;AAChF,IAAA,MAAM,GAAA,GAAM,MAAM,IAAA,CAAK,MAAA,CAAO,YAAY;AACxC,MAAA,OAAO,MAAM,IAAA,CAAK,MAAA,CAAO,QAAA,CAAS;AAAA,QAChC,IAAA;AAAA,QACA,KAAA,EAAO,OAAA;AAAA,QACP,QAAA,EAAU,KAAK,WAAA,EAAa,IAAA;AAAA,QAC5B,MAAA,EAAQ;AAAA,OACT,CAAA;AAAA,IACH,CAAA,EAAG,wBAAwB,CAAA,EAAE;AAE7B,IAAA,OAAO,GAAA;AAAA,EACT;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAOA,MAAM,WAAA,GAAc;AAClB,IAAA,OAAO,EAAE,SAAS,IAAA,EAAK;AAAA,EACzB;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAkBA,MAAM,MAAA,CAAO,KAAA,EAA8B,OAAA,EAAoD;AAC7F,IAAA,MAAM,GAAA,GAAM,MAAM,IAAA,CAAK,MAAA,CAAO,YAAY;AACxC,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,WAAA,MAAiB,SAAS,KAAA,EAAO;AAC/B,QAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,UAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,QAChC,CAAA,MAAO;AACL,UAAA,MAAA,CAAO,KAAK,KAAK,CAAA;AAAA,QACnB;AAAA,MACF;AACA,MAAA,MAAM,MAAA,GAAS,MAAA,CAAO,MAAA,CAAO,MAAM,CAAA;AAEnC,MAAA,MAAM,EAAE,eAAe,gBAAA,EAAkB,YAAA,EAAc,UAAU,GAAG,cAAA,EAAe,GAAI,OAAA,IAAW,EAAC;AAEnG,MAAA,MAAM,IAAA,GAAO,IAAI,IAAA,CAAK,CAAC,MAAM,CAAA,EAAG,CAAA,MAAA,EAAS,QAAA,IAAY,KAAK,CAAA,CAAE,CAAA;AAE5D,MAAA,MAAM,aAAA,GAAgB,MAAM,IAAA,CAAK,MAAA,CAAO,YAAA,CAAa,OAAA;AAAA,QACnD;AAAA,UACE,IAAA;AAAA,UACA,QAAA,EAAU,KAAK,cAAA,EAAgB,IAAA;AAAA,UAC/B,aAAA;AAAA,UACA,gBAAA;AAAA,UACA;AAAA,SACF;AAAA,QACA;AAAA,OACF;AAEA,MAAA,OAAO,aAAA,CAAc,IAAA;AAAA,IACvB,CAAA,EAAG,yBAAyB,CAAA,EAAE;AAE9B,IAAA,OAAO,GAAA;AAAA,EACT;AACF","file":"index.js","sourcesContent":["import { File } from 'node:buffer';\nimport { MastraVoice } from '@mastra/core/voice';\nimport { ElevenLabsClient } from 'elevenlabs';\n\ntype ElevenLabsModel =\n  | 'eleven_multilingual_v2'\n  | 'eleven_flash_v2_5'\n  | 'eleven_flash_v2'\n  | 'eleven_multilingual_sts_v2'\n  | 'eleven_english_sts_v2'\n  | 'scribe_v1';\n\ninterface ElevenLabsVoiceConfig {\n  name?: ElevenLabsModel;\n  apiKey?: string;\n}\n\ninterface SpeechToTextOptions {\n  language_code?: string;\n  tag_audio_events?: boolean;\n  num_speakers?: number;\n  filetype?: string;\n}\n\ninterface RequestOptions {\n  timeoutInSeconds?: number;\n  maxRetries?: number;\n  abortSignal?: AbortSignal;\n  apiKey?: string | undefined;\n  headers?: Record<string, string>;\n}\n\n// Combined options type\ntype ElevenLabsListenOptions = SpeechToTextOptions & RequestOptions;\n\nexport class ElevenLabsVoice extends MastraVoice {\n  private client: ElevenLabsClient;\n\n  /**\n   * Creates an instance of the ElevenLabsVoice class.\n   *\n   * @param {Object} options - The options for the voice configuration.\n   * @param {ElevenLabsVoiceConfig} [options.speechModel] - The configuration for the speech model, including the model name and API key.\n   * @param {string} [options.speaker] - The ID of the speaker to use. If not provided, a default speaker will be used.\n   *\n   * @throws {Error} If the ELEVENLABS_API_KEY is not set in the environment variables.\n   */\n  constructor({\n    speechModel,\n    listeningModel,\n    speaker,\n  }: { speechModel?: ElevenLabsVoiceConfig; listeningModel?: ElevenLabsVoiceConfig; speaker?: string } = {}) {\n    const apiKey = speechModel?.apiKey ?? process.env.ELEVENLABS_API_KEY;\n    super({\n      speechModel: {\n        name: speechModel?.name ?? 'eleven_multilingual_v2',\n        apiKey: speechModel?.apiKey,\n      },\n      listeningModel: {\n        name: listeningModel?.name ?? 'scribe_v1',\n        apiKey: listeningModel?.apiKey,\n      },\n      speaker,\n    });\n\n    if (!apiKey) {\n      throw new Error('ELEVENLABS_API_KEY is not set');\n    }\n\n    this.client = new ElevenLabsClient({\n      apiKey,\n    });\n\n    this.speaker = speaker || '9BWtsMINqrJLrRacOk9x'; // Aria is the default speaker\n  }\n\n  /**\n   * Retrieves a list of available speakers from the Eleven Labs API.\n   * Each speaker includes their ID, name, language, and gender.\n   *\n   * @returns {Promise<Array<{ voiceId: string, name: string, language: string, gender: string }>>}\n   * A promise that resolves to an array of speaker objects.\n   */\n  async getSpeakers() {\n    const res = await this.traced(async () => {\n      const voices = await this.client.voices.getAll();\n      return (\n        voices?.voices?.map(voice => ({\n          voiceId: voice.voice_id,\n          name: voice.name,\n          language: voice.labels?.language || 'en',\n          gender: voice.labels?.gender || 'neutral',\n        })) ?? []\n      );\n    }, 'voice.elevenlabs.voices')();\n\n    return res;\n  }\n\n  private async streamToString(stream: NodeJS.ReadableStream): Promise<string> {\n    const chunks: Buffer[] = [];\n    for await (const chunk of stream) {\n      if (typeof chunk === 'string') {\n        chunks.push(Buffer.from(chunk));\n      } else {\n        chunks.push(chunk);\n      }\n    }\n    return Buffer.concat(chunks).toString('utf-8');\n  }\n\n  /**\n   * Converts text or audio input into speech using the Eleven Labs API.\n   *\n   * @param {string | NodeJS.ReadableStream} input - The text to be converted to speech or a stream containing audio data.\n   * @param {Object} [options] - Optional parameters for the speech generation.\n   * @param {string} [options.speaker] - The ID of the speaker to use for the speech. If not provided, the default speaker will be used.\n   *\n   * @returns {Promise<NodeJS.ReadableStream>} A promise that resolves to a readable stream of the generated speech.\n   *\n   * @throws {Error} If no speaker is specified or if no speech model is set.\n   */\n  async speak(input: string | NodeJS.ReadableStream, options?: { speaker?: string }): Promise<NodeJS.ReadableStream> {\n    const speaker = options?.speaker || this.speaker;\n    if (!speaker) {\n      throw new Error('No speaker specified');\n    }\n\n    if (!this.speechModel?.name) {\n      throw new Error('No speech model specified');\n    }\n    const text = typeof input === 'string' ? input : await this.streamToString(input);\n    const res = await this.traced(async () => {\n      return await this.client.generate({\n        text,\n        voice: speaker,\n        model_id: this.speechModel?.name as ElevenLabsModel,\n        stream: true,\n      });\n    }, 'voice.elevenlabs.speak')();\n\n    return res;\n  }\n\n  /**\n   * Checks if listening capabilities are enabled.\n   *\n   * @returns {Promise<{ enabled: boolean }>}\n   */\n  async getListener() {\n    return { enabled: true };\n  }\n\n  /**\n   * Converts audio input to text using ElevenLabs Speech-to-Text API.\n   *\n   * @param input - A readable stream containing the audio data to transcribe\n   * @param options - Configuration options for the transcription\n   * @param options.language_code - ISO language code (e.g., 'en', 'fr', 'es')\n   * @param options.tag_audio_events - Whether to tag audio events like [MUSIC], [LAUGHTER], etc.\n   * @param options.num_speakers - Number of speakers to detect in the audio\n   * @param options.filetype - Audio file format (e.g., 'mp3', 'wav', 'ogg')\n   * @param options.timeoutInSeconds - Request timeout in seconds\n   * @param options.maxRetries - Maximum number of retry attempts\n   * @param options.abortSignal - Signal to abort the request\n   *\n   * @returns A Promise that resolves to the transcribed text\n   *\n   */\n  async listen(input: NodeJS.ReadableStream, options?: ElevenLabsListenOptions): Promise<string> {\n    const res = await this.traced(async () => {\n      const chunks: Buffer[] = [];\n      for await (const chunk of input) {\n        if (typeof chunk === 'string') {\n          chunks.push(Buffer.from(chunk));\n        } else {\n          chunks.push(chunk);\n        }\n      }\n      const buffer = Buffer.concat(chunks);\n\n      const { language_code, tag_audio_events, num_speakers, filetype, ...requestOptions } = options || {};\n\n      const file = new File([buffer], `audio.${filetype || 'mp3'}`);\n\n      const transcription = await this.client.speechToText.convert(\n        {\n          file: file,\n          model_id: this.listeningModel?.name as ElevenLabsModel,\n          language_code,\n          tag_audio_events,\n          num_speakers,\n        },\n        requestOptions,\n      );\n\n      return transcription.text;\n    }, 'voice.elevenlabs.listen')();\n\n    return res;\n  }\n}\n"]}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@mastra/voice-elevenlabs",
-  "version": "0.10.6",
+  "version": "0.10.7",
   "description": "Mastra ElevenLabs voice integration",
   "type": "module",
   "files": [
@@ -15,7 +15,7 @@
         "default": "./dist/index.js"
       },
       "require": {
-        "types": "./dist/index.d.cts",
+        "types": "./dist/index.d.ts",
         "default": "./dist/index.cjs"
       }
     },
@@ -33,15 +33,16 @@
     "tsup": "^8.5.0",
     "typescript": "^5.8.3",
     "vitest": "^3.2.4",
-    "@internal/lint": "0.0.24",
-    "@mastra/core": "0.12.0"
+    "@internal/lint": "0.0.27",
+    "@mastra/core": "0.13.0",
+    "@internal/types-builder": "0.0.2"
   },
   "peerDependencies": {
-    "@mastra/core": ">=0.10.7-0 <0.13.0-0"
+    "@mastra/core": ">=0.10.7-0 <0.14.0-0"
   },
   "scripts": {
-    "build": "tsup src/index.ts --format esm,cjs --experimental-dts --clean --treeshake=smallest --splitting",
-    "build:watch": "tsup build --watch",
+    "build": "tsup --silent --config tsup.config.ts",
+    "build:watch": "tsup build --watch && tsc -p tsconfig.build.json",
     "test": "vitest run",
     "lint": "eslint ."
   }

package/dist/_tsup-dts-rollup.d.cts

@@ -1,98 +0,0 @@
-import { MastraVoice } from '@mastra/core/voice';
-
-declare type ElevenLabsListenOptions = SpeechToTextOptions & RequestOptions;
-
-declare type ElevenLabsModel = 'eleven_multilingual_v2' | 'eleven_flash_v2_5' | 'eleven_flash_v2' | 'eleven_multilingual_sts_v2' | 'eleven_english_sts_v2' | 'scribe_v1';
-
-export declare class ElevenLabsVoice extends MastraVoice {
-    private client;
-    /**
-     * Creates an instance of the ElevenLabsVoice class.
-     *
-     * @param {Object} options - The options for the voice configuration.
-     * @param {ElevenLabsVoiceConfig} [options.speechModel] - The configuration for the speech model, including the model name and API key.
-     * @param {string} [options.speaker] - The ID of the speaker to use. If not provided, a default speaker will be used.
-     *
-     * @throws {Error} If the ELEVENLABS_API_KEY is not set in the environment variables.
-     */
-    constructor({ speechModel, listeningModel, speaker, }?: {
-        speechModel?: ElevenLabsVoiceConfig;
-        listeningModel?: ElevenLabsVoiceConfig;
-        speaker?: string;
-    });
-    /**
-     * Retrieves a list of available speakers from the Eleven Labs API.
-     * Each speaker includes their ID, name, language, and gender.
-     *
-     * @returns {Promise<Array<{ voiceId: string, name: string, language: string, gender: string }>>}
-     * A promise that resolves to an array of speaker objects.
-     */
-    getSpeakers(): Promise<{
-        voiceId: string;
-        name: string | undefined;
-        language: string;
-        gender: string;
-    }[]>;
-    private streamToString;
-    /**
-     * Converts text or audio input into speech using the Eleven Labs API.
-     *
-     * @param {string | NodeJS.ReadableStream} input - The text to be converted to speech or a stream containing audio data.
-     * @param {Object} [options] - Optional parameters for the speech generation.
-     * @param {string} [options.speaker] - The ID of the speaker to use for the speech. If not provided, the default speaker will be used.
-     *
-     * @returns {Promise<NodeJS.ReadableStream>} A promise that resolves to a readable stream of the generated speech.
-     *
-     * @throws {Error} If no speaker is specified or if no speech model is set.
-     */
-    speak(input: string | NodeJS.ReadableStream, options?: {
-        speaker?: string;
-    }): Promise<NodeJS.ReadableStream>;
-    /**
-     * Checks if listening capabilities are enabled.
-     *
-     * @returns {Promise<{ enabled: boolean }>}
-     */
-    getListener(): Promise<{
-        enabled: boolean;
-    }>;
-    /**
-     * Converts audio input to text using ElevenLabs Speech-to-Text API.
-     *
-     * @param input - A readable stream containing the audio data to transcribe
-     * @param options - Configuration options for the transcription
-     * @param options.language_code - ISO language code (e.g., 'en', 'fr', 'es')
-     * @param options.tag_audio_events - Whether to tag audio events like [MUSIC], [LAUGHTER], etc.
-     * @param options.num_speakers - Number of speakers to detect in the audio
-     * @param options.filetype - Audio file format (e.g., 'mp3', 'wav', 'ogg')
-     * @param options.timeoutInSeconds - Request timeout in seconds
-     * @param options.maxRetries - Maximum number of retry attempts
-     * @param options.abortSignal - Signal to abort the request
-     *
-     * @returns A Promise that resolves to the transcribed text
-     *
-     */
-    listen(input: NodeJS.ReadableStream, options?: ElevenLabsListenOptions): Promise<string>;
-}
-
-declare interface ElevenLabsVoiceConfig {
-    name?: ElevenLabsModel;
-    apiKey?: string;
-}
-
-declare interface RequestOptions {
-    timeoutInSeconds?: number;
-    maxRetries?: number;
-    abortSignal?: AbortSignal;
-    apiKey?: string | undefined;
-    headers?: Record<string, string>;
-}
-
-declare interface SpeechToTextOptions {
-    language_code?: string;
-    tag_audio_events?: boolean;
-    num_speakers?: number;
-    filetype?: string;
-}
-
-export { }

package/dist/_tsup-dts-rollup.d.ts

@@ -1,98 +0,0 @@
-import { MastraVoice } from '@mastra/core/voice';
-
-declare type ElevenLabsListenOptions = SpeechToTextOptions & RequestOptions;
-
-declare type ElevenLabsModel = 'eleven_multilingual_v2' | 'eleven_flash_v2_5' | 'eleven_flash_v2' | 'eleven_multilingual_sts_v2' | 'eleven_english_sts_v2' | 'scribe_v1';
-
-export declare class ElevenLabsVoice extends MastraVoice {
-    private client;
-    /**
-     * Creates an instance of the ElevenLabsVoice class.
-     *
-     * @param {Object} options - The options for the voice configuration.
-     * @param {ElevenLabsVoiceConfig} [options.speechModel] - The configuration for the speech model, including the model name and API key.
-     * @param {string} [options.speaker] - The ID of the speaker to use. If not provided, a default speaker will be used.
-     *
-     * @throws {Error} If the ELEVENLABS_API_KEY is not set in the environment variables.
-     */
-    constructor({ speechModel, listeningModel, speaker, }?: {
-        speechModel?: ElevenLabsVoiceConfig;
-        listeningModel?: ElevenLabsVoiceConfig;
-        speaker?: string;
-    });
-    /**
-     * Retrieves a list of available speakers from the Eleven Labs API.
-     * Each speaker includes their ID, name, language, and gender.
-     *
-     * @returns {Promise<Array<{ voiceId: string, name: string, language: string, gender: string }>>}
-     * A promise that resolves to an array of speaker objects.
-     */
-    getSpeakers(): Promise<{
-        voiceId: string;
-        name: string | undefined;
-        language: string;
-        gender: string;
-    }[]>;
-    private streamToString;
-    /**
-     * Converts text or audio input into speech using the Eleven Labs API.
-     *
-     * @param {string | NodeJS.ReadableStream} input - The text to be converted to speech or a stream containing audio data.
-     * @param {Object} [options] - Optional parameters for the speech generation.
-     * @param {string} [options.speaker] - The ID of the speaker to use for the speech. If not provided, the default speaker will be used.
-     *
-     * @returns {Promise<NodeJS.ReadableStream>} A promise that resolves to a readable stream of the generated speech.
-     *
-     * @throws {Error} If no speaker is specified or if no speech model is set.
-     */
-    speak(input: string | NodeJS.ReadableStream, options?: {
-        speaker?: string;
-    }): Promise<NodeJS.ReadableStream>;
-    /**
-     * Checks if listening capabilities are enabled.
-     *
-     * @returns {Promise<{ enabled: boolean }>}
-     */
-    getListener(): Promise<{
-        enabled: boolean;
-    }>;
-    /**
-     * Converts audio input to text using ElevenLabs Speech-to-Text API.
-     *
-     * @param input - A readable stream containing the audio data to transcribe
-     * @param options - Configuration options for the transcription
-     * @param options.language_code - ISO language code (e.g., 'en', 'fr', 'es')
-     * @param options.tag_audio_events - Whether to tag audio events like [MUSIC], [LAUGHTER], etc.
-     * @param options.num_speakers - Number of speakers to detect in the audio
-     * @param options.filetype - Audio file format (e.g., 'mp3', 'wav', 'ogg')
-     * @param options.timeoutInSeconds - Request timeout in seconds
-     * @param options.maxRetries - Maximum number of retry attempts
-     * @param options.abortSignal - Signal to abort the request
-     *
-     * @returns A Promise that resolves to the transcribed text
-     *
-     */
-    listen(input: NodeJS.ReadableStream, options?: ElevenLabsListenOptions): Promise<string>;
-}
-
-declare interface ElevenLabsVoiceConfig {
-    name?: ElevenLabsModel;
-    apiKey?: string;
-}
-
-declare interface RequestOptions {
-    timeoutInSeconds?: number;
-    maxRetries?: number;
-    abortSignal?: AbortSignal;
-    apiKey?: string | undefined;
-    headers?: Record<string, string>;
-}
-
-declare interface SpeechToTextOptions {
-    language_code?: string;
-    tag_audio_events?: boolean;
-    num_speakers?: number;
-    filetype?: string;
-}
-
-export { }

package/dist/index.d.cts

@@ -1 +0,0 @@
-export { ElevenLabsVoice } from './_tsup-dts-rollup.cjs';