> ## Documentation Index
> Fetch the complete documentation index at: https://docs.geminix.cc/llms.txt
> Use this file to discover all available pages before exploring further.

# 音频转录

> 将音频转换为文本



## OpenAPI

````yaml /openapi/relay.json post /v1/audio/transcriptions
openapi: 3.0.1
info:
  title: OmniMux Gateway API
  description: >-
    OpenAI-compatible relay API for OmniMux (geminix.cc). Use Authorization:
    Bearer sk-... and base URL https://api.geminix.cc/v1 for SDKs.
  version: 1.0.0
servers:
  - url: https://api.geminix.cc
    description: OmniMux production API
security:
  - BearerAuth: []
tags:
  - name: 获取模型列表
  - name: OpenAI格式(Chat)
  - name: OpenAI格式(Responses)
  - name: 图片生成
  - name: 图片生成/OpenAI兼容格式
  - name: 图片生成/Qwen千问
  - name: 视频生成
  - name: 视频生成/Sora兼容格式
  - name: 视频生成/Kling格式
  - name: 视频生成/即梦格式
  - name: Claude格式(Messages)
  - name: Gemini格式
  - name: OpenAI格式(Embeddings)
  - name: 文本补全(Completions)
  - name: OpenAI音频(Audio)
  - name: 重排序(Rerank)
  - name: Moderations
  - name: Realtime
  - name: 未实现
  - name: 未实现/Fine-tunes
  - name: 未实现/Files
paths:
  /v1/audio/transcriptions:
    post:
      tags:
        - Audio
      summary: 音频转录
      description: 将音频转换为文本
      operationId: createTranscription
      parameters: []
      requestBody:
        content:
          multipart/form-data:
            schema:
              type: object
              properties:
                file:
                  type: string
                  format: binary
                  description: 音频文件
                  example: ''
                model:
                  type: string
                  example: whisper-1
                language:
                  type: string
                  description: ISO-639-1 语言代码
                  example: ''
                prompt:
                  type: string
                  example: ''
                response_format:
                  type: string
                  enum:
                    - json
                    - text
                    - srt
                    - verbose_json
                    - vtt
                  default: json
                  example: json
                temperature:
                  type: number
                  example: 0
                timestamp_granularities:
                  type: array
                  items:
                    type: string
                    enum:
                      - word
                      - segment
                  example: ''
              required:
                - file
                - model
        required: true
      responses:
        '200':
          description: 成功转录
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/AudioTranscriptionResponse'
          headers: {}
      deprecated: false
      security:
        - BearerAuth: []
components:
  schemas:
    AudioTranscriptionResponse:
      type: object
      properties:
        text:
          type: string
  securitySchemes:
    BearerAuth:
      type: http
      scheme: bearer
      description: |
        使用 Bearer Token 认证。
        格式: `Authorization: Bearer sk-xxxxxx`

````