Pronunciation & Voice Coach

Command

Environment

• MCP_PRONUNCIATION_MODEL

  faster-whisper model size. Default: base.en. Options: tiny.en, base.en, small.en, medium.en, large-v3, large-v3-turbo.

• HF_HUB_CACHE

  Override Hugging Face Hub cache directory where Whisper weights are stored.

• MCP_PRONUNCIATION_PRELOAD

  Set to 0, false, no, or off to disable background Whisper model preload during registry inspection or Docker smoke tests.

• MCP_PRONUNCIATION_AUDIO_RETENTION

  Temporary WAV retention policy. Use session to delete recordings when the server exits, or keep for manual inspection.

• TORCH_HOME

  Override PyTorch cache directory for optional wav2vec2 forced-alignment weights.