POST
/
audio
/
speech
curl --request POST \
  --url https://api.siliconflow.cn/v1/audio/speech \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "fishaudio/fish-speech-1.5",
  "input": "The text to generate audio for",
  "voice": "fishaudio/fish-speech-1.5:alex",
  "response_format": "mp3",
  "sample_rate": 32000,
  "stream": true,
  "speed": 1,
  "gain": 0
}'
This response does not have an example.

Authorizations

​
Authorization
string
header
required

Use the following format for authentication: Bearer <your api key>

Body

application/json
​
model
enum<string>
required

对应的模型名称。为更好的提升服务质量，我们将不定期对本服务提供的模型做相关变更，包括但不限于模型上下线，价格调整，模型服务能力调整等，我们会在可行的情况下以公告、消息推送等适当的方式进行通知。

Available options:
fishaudio/fish-speech-1.5
​
input
string
required

The text to generate audio for.

Required string length: 1 - 128000
​
voice
enum<string>
required
Available options:
fishaudio/fish-speech-1.5:alex,
fishaudio/fish-speech-1.5:anna,
fishaudio/fish-speech-1.5:bella,
fishaudio/fish-speech-1.5:benjamin,
fishaudio/fish-speech-1.5:charles,
fishaudio/fish-speech-1.5:claire,
fishaudio/fish-speech-1.5:david,
fishaudio/fish-speech-1.5:diana
​
response_format
enum<string>
default:
mp3

The format to audio out. Supported formats are mp3, opus, wav, pcm

Available options:
mp3,
opus,
wav,
pcm
​
sample_rate
enum<number>

控制输出采样率，对于不同的视频输出类型，默认值和可取值范围不同，具体如下：1. opus: 目前支持48000 Hz 2. wav, pcm: 支持8000, 16000, 24000, 32000, 44100 Hz, 默认44100 Hz 3. mp3: 支持32000, 44100 Hz, 默认44100 Hz

Available options:
8000,
16000,
24000,
32000,
44100,
48000
​
stream
boolean
default:
true

streaming or not

​
speed
number
default:
1

The speed of the generated audio. Select a value from 0.25 to 4.0. 1.0 is the default.

Required range: 0.25 < x < 4
​
gain
number
default:
0
Required range: -10 < x < 10

Response

200
application/audio
根据输入的文本生成音频。接口生成的数据为音频的二进制数据，需要使用者自行处理。参考：https://docs.siliconflow.cn/capabilities/text-to-speech#5

The response is of type file.

上传参考音频创建语音转文本请求