Whisper Tiny it 11
This model is a fine-tuned version of openai/whisper-tiny on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.670211
- Wer: 42.276761
Model description
This model is the openai whisper small transformer adapted for Italian audio to text transcription.
Intended uses & limitations
The model is available through its HuggingFace web app
Training and evaluation data
Data used for training is the initial 25% of train and validation of Italian Common Voice 11.0 from Mozilla Foundation. The dataset used for evaluation is the initial 10% of test of Italian Common Voice.
Training procedure
After loading the pre trained model, it has been trained on the dataset.
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 16
- eval_batch_size: 8
- 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: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.584600 | 0.95 | 1000 | 0.801204 | 48.980865 |
| 0.496100 | 1.91 | 2000 | 0.713927 | 46.283971 |
| 0.406000 | 2.86 | 3000 | 0.680141 | 43.268164 |
| 0.402000 | 3.82 | 4000 | 0.670211 | 42.276761 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2
