distil-whisper/distil-large-v3.5-ggml

Distil-Whisper: Distil-Large-v3.5 for Whisper.cpp

This repository contains the model weights for distil-large-v3.5 converted to GGML format. GGML is the weight format expected by C/C++ packages such as Whisper.cpp, for which we provide an example below.

Usage

Distil-Whisper-v3.5 can be run with the Whisper.cpp package with the original sequential long-form transcription algorithm.

Steps for getting started:

1. Clone and build the Whisper.cpp repository:

git clone https://github.com/ggerganov/whisper.cpp.git
cd whisper.cpp
make

2. Install the Hugging Face Hub Python package:

pip install --upgrade huggingface_hub

And download the GGML weights for distil-large-v3 using the following Python snippet:

from huggingface_hub import hf_hub_download

hf_hub_download(repo_id='distil-whisper/distil-large-v3.5-ggml', filename='ggml-model.bin', local_dir='./models')

Note that if you do not have a Python environment set-up, you can also download the weights directly with wget:

wget https://huggingface.co/distil-whisper/distil-large-v3.5-ggml/resolve/main/ggml-model.bin -P ./models

3. Run inference using the provided sample audio:

./main -m ./models/ggml-model.bin -l en -f /path/to/audio/file --print-colors

Model Details

For more information about the Distil-Large-v3.5 model, refer to the original model card.

License

Distil-Whisper inherits the MIT license from OpenAI's Whisper model.

Citation

If you use this model, please consider citing the Distil-Whisper paper:

@misc{gandhi2023distilwhisper,
      title={Distil-Whisper: Robust Knowledge Distillation via Large-Scale Pseudo Labelling}, 
      author={Sanchit Gandhi and Patrick von Platen and Alexander M. Rush},
      year={2023},
      eprint={2311.00430},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}