FasterWhisper fine tuned models by Xabier de Zuazo for all the iberian peninsula languages, pt, es, gl, ca, eu

NOTE: this is intended to use only these specific finetuned models, if you want to use your own models use the main ovos-stt-plugin-fasterwhisper instead

pip install ovos-stt-plugin-fasterwhisper-zuazo

to use small model locally

  "stt": {
    "module": "ovos-stt-plugin-fasterwhisper-zuazo",
    "ovos-stt-plugin-fasterwhisper-zuazo": {
        "model": "small"
    }
  }

to use Large model with GPU

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

best model is automatically selected based on language unless specified, valid config options are tiny,base,small,medium,large

you can also select large-v1, large-v2 or large-v3 explicitly, but v3 is only available for pt,ca,eu

self reported WER score from model pages

self reported WER score from model pages

self reported WER score from model pages

self reported WER score from model pages

self reported WER score from model pages

This plugin was funded by the Ministerio para la Transformación Digital y de la Función Pública and Plan de Recuperación, Transformación y Resiliencia - Funded by EU – NextGenerationEU within the framework of the project ILENIA with reference 2022/TL22/00215337