Skip to content

数据模型: AudioTranscriptionRequest ​

AudioTranscriptionRequest ​

字段类型必选描述
filestring是音频文件
modelstring是
languagestring否ISO-639-1 语言代码
promptstring否
response_formatstring否
temperaturenumber否
timestamp_granularitiesarray of string否

数据模型示例 ​

json
{
  "file": "string",
  "model": "whisper-1",
  "language": "string",
  "prompt": "string",
  "response_format": "string",
  "temperature": 0,
  "timestamp_granularities": [
    "string"
  ]
}

Released under the MIT License.