> ## Documentation Index > Fetch the complete documentation index at: https://docs.siliconflow.cn/llms.txt > Use this file to discover all available pages before exploring further. # 创建文本转语音请求 > Generate audio from input text. The data generated by the interface is the binary data of the audio, which requires the user to handle it themselves. Reference:https://docs.siliconflow.cn/capabilities/text-to-speech#5 ## OpenAPI ````yaml post /audio/speech openapi: 3.0.0 info: title: SiliconFlow API description: The SiliconFlow REST API version: 1.0.0 contact: name: SiliconFlow Support url: https://www.siliconflow.cn/ license: name: MIT url: https://github.com/siliconflow/siliconcloud/blob/main/LICENSE servers: - url: https://api.siliconflow.cn/v1 security: - bearerAuth: [] paths: /audio/speech: post: tags: - Audio summary: Create Speech description: >- Generate audio from input text. The data generated by the interface is the binary data of the audio, which requires the user to handle it themselves. Reference:https://docs.siliconflow.cn/capabilities/text-to-speech#5 operationId: createSpeech requestBody: required: true content: application/json: schema: oneOf: - $ref: '#/components/schemas/MOSS-TTSD-v0.5' - $ref: '#/components/schemas/CosyVoice2-0.5B' responses: '200': description: >- Generate audio based on the input text. The data generated by the interface is in binary format and requires the user to process it themselves. Reference:https://docs.siliconflow.cn/capabilities/text-to-speech#5 The response header contains the x-siliconcloud-trace-id field, which serves as a unique identifier for tracing requests, facilitating log queries and issue troubleshooting. headers: Transfer-Encoding: schema: type: string description: chunked content: application/audio: schema: type: string format: binary example: 音频的二进制数据 audio/wav: schema: type: string format: binary example: 音频的二进制数据 audio/opus: schema: type: string format: binary example: 音频的二进制数据 '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '403': $ref: '#/components/responses/Forbidden' '404': $ref: '#/components/responses/NotFound' '429': $ref: '#/components/responses/RateLimit' '503': $ref: '#/components/responses/Overloaded' '504': $ref: '#/components/responses/Timeout' x-codeSamples: - lang: python label: Default source: > import requests url = "https://api.siliconflow.cn/v1/audio/speech" payload = { "model": "fnlp/MOSS-TTSD-v0.5", "input": "你站在桥上看风景，看风景的人在楼上看你。明月装饰了你的窗子，你装饰了别人的梦", "voice": "fnlp/MOSS-TTSD-v0.5:alex", "response_format": "mp3", "stream": True } headers = { "Authorization": "Bearer sk-xx", # 请替换为您的真实 API Key "Content-Type": "application/json" } response = requests.post(url, json=payload, headers=headers, stream=True) if response.status_code == 200: # 将流式响应写入文件 with open("output.mp3", "wb") as f: for chunk in response.iter_content(chunk_size=1024): if chunk: f.write(chunk) print("音频已成功保存为 output.mp3") else: print(f"请求失败，状态码: {response.status_code}") print(f"错误信息: {response.text}") - lang: curl label: Default source: | curl --location 'https://api.siliconflow.cn/v1/audio/speech' \ --header 'Authorization: Bearer sk-xx' \ --header 'Content-Type: application/json' \ --data '{ "model": "fnlp/MOSS-TTSD-v0.5", "input": "你站在桥上看风景，看风景的人在楼上看你。明月装饰了你的窗子，你装饰了别人的梦", "voice": "fnlp/MOSS-TTSD-v0.5:alex", "response_format": "mp3", "stream": true }' - lang: javaScript label: Default source: | const axios = require('axios'); const fs = require('fs'); const url = 'https://api.siliconflow.cn/v1/audio/speech'; const data = { model: "fnlp/MOSS-TTSD-v0.5", input: "你站在桥上看风景，看风景的人在楼上看你。明月装饰了你的窗子，你装饰了别人的梦", voice: "fnlp/MOSS-TTSD-v0.5:alex", response_format: "mp3", stream: true }; const config = { method: 'post', url: url, headers: { 'Authorization': 'Bearer sk-xx', // 请替换为您的真实 API Key 'Content-Type': 'application/json' }, data: data, responseType: 'stream' // 重要：设置响应类型为流 }; axios(config) .then(function (response) { // 将流式数据写入文件 const writer = fs.createWriteStream('output.mp3'); response.data.pipe(writer); writer.on('finish', () => { console.log('音频已成功保存为 output.mp3'); }); writer.on('error', (err) => { console.error('写入文件时出错:', err); }); }) .catch(function (error) { console.error('请求失败:', error.message); if (error.response) { console.error('状态码:', error.response.status); } }); components: schemas: MOSS-TTSD-v0.5: title: MOSS-TTSD-v0.5 type: object required: - model - input additionalProperties: false properties: model: type: string enum: - fnlp/MOSS-TTSD-v0.5 description: > MOSS-TTSD (text to spoken dialogue) is an open-source bilingual spoken dialogue synthesis model that supports both Chinese and English. It can transform dialogue scripts between two speakers into natural, expressive conversational speech. MOSS-TTSD supports voice cloning and long single-session speech generation, making it ideal for AI podcast production. To better enhance service quality, we will make periodic changes to the models provided by this service, including but not limited to model on/offlining and adjustments to model service capabilities. We will notify you of such changes through appropriate means such as announcements or message pushes where feasible. input: type: string description: | The dialogue text uses speaker tags to indicate turns: [S1]: Indicates Speaker 1 is speaking [S2]: Indicates Speaker 2 is speaking example: >- [S1]Hello, how are you today?[S2]I'm doing great, thanks for asking![S1]That's wonderful to hear maxLength: 128000 minLength: 1 max_tokens: type: integer description: >- The maximum number of tokens to generate. The input + output does not exceed 32k tokens. default: 2048 example: 4096 references: description: >- The voice field and references field are mutually exclusive. If you want to use scripted dialogue, you need to pass two voice tones through the references field. Scripted dialogue is only available for the moss model. type: array items: type: object properties: audio: oneOf: - type: string format: url description: >- A URL pointing to an audio file (e.g., `https://example.com/audio.mp3`). - type: string pattern: ^data:audio\/\w+;base64,[A-Za-z0-9+/=]+$ description: >- A base64-encoded audio string (e.g., `data:audio/mp3;base64,ABC123...`). text: description: >- The audio content, which can be either a URL pointing to an audio file or a base64-encoded audio string. type: string voice: description: >- The "voice" field currently does not support two timbres. If you need to upload two timbres, please use "reference". type: string enum: - fnlp/MOSS-TTSD-v0.5:alex - fnlp/MOSS-TTSD-v0.5:anna - fnlp/MOSS-TTSD-v0.5:bella - fnlp/MOSS-TTSD-v0.5:benjamin - fnlp/MOSS-TTSD-v0.5:charles - fnlp/MOSS-TTSD-v0.5:claire - fnlp/MOSS-TTSD-v0.5:david - fnlp/MOSS-TTSD-v0.5:diana response_format: description: >- The format to audio out. Supported formats are `mp3`, `opus`, `wav`, `pcm` default: mp3 type: string enum: - mp3 - opus - wav - pcm sample_rate: description: >- Control the output sample rate. The default values and differ for different video output types, as follows: opus: Supports 48000 Hz. wav, pcm: Supports 8000, 16000, 24000, 32000, 44100 Hz, with a default of 44100 Hz. mp3: Supports 32000, 44100 Hz, with a default of 44100 Hz. type: number default: 32000 stream: description: streaming or not type: boolean default: true speed: type: number description: >- The speed of the generated audio. Select a value from `0.25` to `4.0`. `1.0` is the default. format: float minimum: 0.25 maximum: 4 default: 1 gain: type: number format: float minimum: -10 maximum: 10 default: 0 CosyVoice2-0.5B: title: CosyVoice2-0.5B type: object required: - model - input additionalProperties: false properties: model: type: string enum: - FunAudioLLM/CosyVoice2-0.5B description: >- Corresponding Model Name. To better enhance service quality, we will make periodic changes to the models provided by this service, including but not limited to model on/offlining and adjustments to model service capabilities. We will notify you of such changes through appropriate means such as announcements or message pushes where feasible. input: type: string description: >- For natural language instructions, add a special end marker "<|endofprompt|>" before the natural language description. These descriptions cover aspects such as emotion, speaking speed, role-playing, and dialects. For detailed instructions, insert pitch bursts between text markers, using markers like "[laughter]" and "[breath]." Additionally, we apply pitch feature markers to phrases; for example:Can you say it with a happy emotion? <|endofprompt|> Today is really happy, Spring Festival is coming! I’m so happy, Spring Festival is coming! [laughter] [breath]. example: >- Can you say it with a happy emotion? <|endofprompt|>I'm so happy, Spring Festival is coming! default: >- Can you say it with a happy emotion? <|endofprompt|>I'm so happy, Spring Festival is coming! maxLength: 128000 minLength: 1 voice: type: string enum: - FunAudioLLM/CosyVoice2-0.5B:alex - FunAudioLLM/CosyVoice2-0.5B:anna - FunAudioLLM/CosyVoice2-0.5B:bella - FunAudioLLM/CosyVoice2-0.5B:benjamin - FunAudioLLM/CosyVoice2-0.5B:charles - FunAudioLLM/CosyVoice2-0.5B:claire - FunAudioLLM/CosyVoice2-0.5B:david - FunAudioLLM/CosyVoice2-0.5B:diana references: description: The voice field and references field are mutually exclusive. type: array items: type: object properties: audio: oneOf: - type: string format: uri description: >- A URL pointing to an audio file (e.g., `https://example.com/audio.mp3`). - type: string pattern: ^data:audio\/\w+;base64,[A-Za-z0-9+/=]+$ description: >- A base64-encoded audio string (e.g., `data:audio/mp3;base64,ABC123...`). text: description: >- The audio content, which can be either a URL pointing to an audio file or a base64-encoded audio string. type: string response_format: description: >- The format to audio out. Supported formats are `mp3`, `opus`, `wav`, `pcm` default: mp3 type: string enum: - mp3 - opus - wav - pcm sample_rate: description: >- Control the output sample rate. The default values and differ for different video output types, as follows: opus: Supports 48000 Hz. wav, pcm: Supports 8000, 16000, 24000, 32000, 44100 Hz, with a default of 44100 Hz. mp3: Supports 32000, 44100 Hz, with a default of 44100 Hz. type: number default: 32000 stream: description: streaming or not type: boolean speed: type: number description: >- The speed of the generated audio. Select a value from `0.25` to `4.0`. `1.0` is the default. format: float minimum: 0.25 maximum: 4 default: 1 gain: type: number format: float minimum: -10 maximum: 10 default: 0 BadRquestData: type: object required: - message - data - code properties: code: type: integer nullable: true default: false example: 20012 message: type: string nullable: false data: type: string nullable: false UnauthorizedData: type: string default: false example: Invalid token ForbiddenData: type: string default: false example: Forbidden NotFoundData: type: string default: false example: 404 page not found RateLimitData: type: object required: - message - data properties: message: type: string example: >- Request was rejected due to rate limiting. If you want more, please contact contact@siliconflow.cn. Details:TPM limit reached. data: type: string OverloadedtData: type: object required: - code - message - data properties: code: type: integer example: 50505 message: type: string example: Model service overloaded. Please try again later. data: type: string nullable: false TimeoutData: type: string responses: BadRequest: description: BadRequest content: application/json: schema: $ref: '#/components/schemas/BadRquestData' Unauthorized: description: Unauthorized content: application/json: schema: $ref: '#/components/schemas/UnauthorizedData' Forbidden: description: Forbidden content: application/json: schema: $ref: '#/components/schemas/ForbiddenData' NotFound: description: NotFound content: application/json: schema: $ref: '#/components/schemas/NotFoundData' RateLimit: description: RateLimit content: application/json: schema: $ref: '#/components/schemas/RateLimitData' Overloaded: description: Overloaded content: application/json: schema: $ref: '#/components/schemas/OverloadedtData' Timeout: description: Timeout content: application/json: schema: $ref: '#/components/schemas/TimeoutData' securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: your api key description: >- Use the following format for authentication: Bearer [](https://cloud.siliconflow.cn/account/ak) ````