Whisper Small3 Italian

This model is a fine-tuned version of openai/whisper-small on the mozilla-foundation/common_voice_11_0 it dataset. It achieves the following results on the evaluation set:

Loss: 0.2307
Wer: 10.2508

Model description

This model is a fine-tuning of the OpenAI Whisper Small model, on the specified dataset.

Intended uses & limitations

This model has been developed as part of the Hugging Face Whisper Fine Tuning sprint, December 2022.

It is meant to spread the knowledge on how these models are built and can be used to develop solutions where it is needed ASR on the Italian Language.

It has not been extensively tested. It is possible that on other datasets the accuracy will be lower.

Please, test it before using it.

Training and evaluation data

Trained and tested on Mozilla Common Voice, vers. 11

Training procedure

The script run.sh, and the Python file, used for the training are saved in the repository.

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 8e-06
train_batch_size: 64
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
training_steps: 6000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.226	2.01	1000	0.2494	11.3684
0.1017	4.02	2000	0.2403	10.6029
0.0491	6.03	3000	0.2549	10.9591
0.1102	8.04	4000	0.2307	10.2508
0.0384	10.05	5000	0.2592	10.5903
0.0285	12.06	6000	0.2537	10.5026

Framework versions

Transformers 4.26.0.dev0
Pytorch 1.13.0+cu117
Datasets 2.7.1.dev0
Tokenizers 0.13.2

luigisaetta
/

whisper-small3-it