OpenVoiceOS STT plugin for Faster Whisper

High-performance inference of OpenAI's Whisper automatic speech recognition (ASR) model:

pip install ovos-stt-plugin-fasterwhisper

available models are 'tiny.en', 'tiny', 'base.en', 'base', 'small.en', 'small', 'medium.en', 'medium', 'large-v1', 'large-v2', 'large-v3', 'large', 'distil-large-v2', 'distil-medium.en', 'distil-small.en', 'distil-large-v3'

you can also pass a full path to a local model or a huggingface repo_id, eg. "projecte-aina/faster-whisper-large-v3-ca-3catparla"

You can convert any whisper model, or use any compatible model from huggingface

to use Large model with GPU

  "stt": {
    "module": "ovos-stt-plugin-fasterwhisper",
    "ovos-stt-plugin-fasterwhisper": {
        "model": "large-v3",
        "use_cuda": true,
        "compute_type": "float16",
        "beam_size": 5,
        "cpu_threads": 4
    }
  }

To use Whisper for lang detection (ovos-dinkum-listener only)

  "listener": {
    "audio_transformers": {
        "ovos-audio-transformer-plugin-fasterwhisper": {
            "model": "small"
        }
    }
  }