faster-whisper-large-v3-ca-punctuated-3370h

Table of Contents

Click to expand

• Model Description
• Intended Uses and Limitations
• How to Get Started with the Model
• Conversion Details
• Citation
• Additional information

Summary

The "faster-whisper-large-v3-ca-punctuated-3370h" is an acoustic model based on a faster-whisper version of langtech-veu/whisper-large-v3-ca-punctuated-3370h suitable for Automatic Speech Recognition in Catalan.

Model Description

The "faster-whisper-large-v3-ca-punctuated-3370h" is the result of converting the langtech-veu/whisper-large-v3-ca-punctuated-3370h into a lighter model using a Python module called faster-whisper.

Intended Uses and Limitations

This model can be used for Automatic Speech Recognition (ASR) in Catalan. The model intends to transcribe Catalan audio files to plain text with punctuation.

How to Get Started with the Model

To see an updated and functional version of this code, please visit our Notebook.

Installation

To use this model, you may install faster-whisper

Create a virtual environment:

python -m venv /path/to/venv

Activate the environment:

source /path/to/venv/bin/activate

Install the modules:

pip install faster-whisper

For Inference

To transcribe audio in Catalan using this model, you can follow this example:

from faster_whisper import WhisperModel

model_size = "langtech-veu/faster-whisper-large-v3-ca-punctuated-3370h"

# Run on GPU with FP16
model = WhisperModel(model_size, device="cuda", compute_type="float16")

# or run on GPU with INT8
#model = WhisperModel(model_size, device="cuda", compute_type="int8_float16")
# or run on CPU with INT8
# model = WhisperModel(model_size, device="cpu", compute_type="int8")

segments, info = model.transcribe("audio_in_catalan.mp3", beam_size=5, task="transcribe",language="ca")

print("Detected language '%s' with probability %f" % (info.language, info.language_probability))

for segment in segments:
    print("[%.2fs -> %.2fs] %s" % (segment.start, segment.end, segment.text))

Conversion Details

Conversion procedure

This model is not a direct result of training. It is a conversion of a Whisper model using faster-whisper. The procedure to create the model is as follows:

ct2-transformers-converter --model langtech-veu/whisper-large-v3-ca-punctuated-3370h
   --output_dir faster-whisper-large-v3-ca-punctuated-3370h
   --copy_files preprocessor_config.json 
   --quantization float16

Citation

If this model contributes to your research, please cite the work:

@misc{mena2025whisperpunctuated,
      title={Acoustic Model in Catalan: whisper-large-v3-ca-punctuated-3370h.}, 
      author={Hernandez Mena, Carlos Daniel, Messaoudi, Abir, España-Bonet, Cristina},
      organization={Barcelona Supercomputing Center},
      url={https://huggingface.co/langtech-veu/faster-whisper-large-v3-ca-punctuated-3370h},
      year={2025}
}

Additional Information

Author

The conversion process was performed during May (2025) in the Language Technologies Laboratory of the Barcelona Supercomputing Center by Abir Messaoudi.

Contact

For further information, please send an email to [email protected].

Copyright

Copyright(c) 2025 by Language Technologies Laboratory, Barcelona Supercomputing Center.

License

Apache-2.0

Funding

This work has been promoted and financed by the Generalitat de Catalunya through the Aina project.

The conversion of the model was possible thanks to the computing time provided by Barcelona Supercomputing Center through MareNostrum 5.