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

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

Table of Contents

<details> <summary>Click to expand</summary>

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

</details>

Model Description

The "whisper-large-v3-ca-punctuated-3370h" is an acoustic model suitable for Automatic Speech Recognition in Catalan. It is the result of finetuning the model "openai/whisper-large-v3" with a combination of Catalan data from Common Voice 17.0 (2,659 hours) and 710 hours of data released by the Projecte AINA from Barcelona, Spain. Totalling 3369 hours and 53 minutes.

A key advantage of this model is that it was trained on meticulously transcribed data, including punctuation and capitalization. As a result, the output transcriptions preserve these features, delivering more structured and readable outputs compared to standard ASR models.

Intended Uses and Limitations

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

How to Get Started with the Model

To see a functional version of this code, please see our our Notebook and, in order to invoke this model, just substitute the instances of "projecte-aina/whisper-large-v3-ca-3catparla" with "langtech-veu/whisper-large-v3-ca-punctuated-3370h".

Installation

In order to use this model, you may install datasets and transformers:

Create a virtual environment:

python -m venv /path/to/venv

Activate the environment:

source /path/to/venv/bin/activate

Install the modules:

pip install datasets transformers 

For Inference

In order to transcribe audio in Catalan using this model, you can follow this example:

#Install Prerequisites
pip install torch
pip install datasets
pip install 'transformers[torch]'
pip install evaluate
pip install jiwer

#This code works with GPU

#Notice that: load_metric is no longer part of datasets.
#you have to remove it and use evaluate's load instead.
#(Note from November 2024)

import torch
from transformers import WhisperForConditionalGeneration, WhisperProcessor

#Load the processor and model.
MODEL_NAME="BSC-LT/whisper-large-v3-ca-punctuated-3370h"
processor = WhisperProcessor.from_pretrained(MODEL_NAME)
model = WhisperForConditionalGeneration.from_pretrained(MODEL_NAME).to("cuda")

#Load the dataset
from datasets import load_dataset, load_metric, Audio
ds=load_dataset("projecte-aina/parlament_parla",split='test')

#Downsample to 16kHz
ds = ds.cast_column("audio", Audio(sampling_rate=16_000))

#Process the dataset
def map_to_pred(batch):
	audio = batch["audio"]
	input_features = processor(audio["array"], sampling_rate=audio["sampling_rate"], return_tensors="pt").input_features
	batch["reference"] = processor.tokenizer._normalize(batch['normalized_text'])

	with torch.no_grad():
		predicted_ids = model.generate(input_features.to("cuda"))[0]
	
	transcription = processor.decode(predicted_ids)
	batch["prediction"] = processor.tokenizer._normalize(transcription)
	
	return batch
	
#Do the evaluation
result = ds.map(map_to_pred)

#Compute the overall WER now.
from evaluate import load

wer = load("wer")
WER=100 * wer.compute(references=result["reference"], predictions=result["prediction"])
print(WER)

<!-- **Test Result**: 0.96 -->

Training Details

Training data

The specific datasets used to create the model are:

• Common Voice 17.0
• "3CatParla". (soon to be published)

Training procedure

This model is the result of finetuning the model "openai/whisper-large-v3" by following this tutorial provided by Hugging Face.

Training Hyperparameters

• language: catalan
• hours of training audio: 3369 hours and 53 minutes
• learning rate: 1e-5
• sample rate: 16000
• train batch size: 32 (x4 GPUs)
  • gradient accumulation steps: 1
• eval batch size: 32
• save total limit: 4
• max steps: 77660
• warmup steps: 7766
• eval steps: 7766
• save steps: 7766

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},
      organization={Barcelona Supercomputing Center},
      url={https://huggingface.co/langtech-veu/whisper-large-v3-ca-punctuated-3370h},
      year={2025}
}

Additional Information

Author

The fine-tuning process was performed during April (2025) in the Language Technologies Laboratory of the Barcelona Supercomputing Center by Carlos Daniel Hernández Mena.

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