docker-compose.yml

version: '3.8'
services:
  speech-to-text:
    image: speech-to-text:latest
    container_name: speech-to-text
    environment:
      - S2T_API_URL=your-speech-to-text-api-endpoint-url
      # 语音转录 API 地址，可选配置，默认是硅基(https://api.siliconflow.cn/v1/audio/transcriptions)
      - S2T_API_KEY=your-speech-to-text-api-key
      # 语音转录 API Key，必须配置
      - S2T_MODEL=your-speech-to-text-model
      # 语音转录模型，可选配置，默认是 FunAudioLLM/SenseVoiceSmall
      - OPT_API_URL=your-text-optimizing-api-endpoint-url
      # 文本校准优化 API 地址，可选配置，不配置不启用优化，直接转出转录结果
      - OPT_API_KEY=your-text-optimizing-api-key
      # 文本校准优化 API Key
      - OPT_MODEL=your-text-optimizing-model
      # 文本校准优化模型
      - API_ACCESS_TOKEN=your-api-auth-key
      # API 密钥
    ports:
      - "your-port:5000"