metadata
library_name: transformers
language:
- it
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_4_0
metrics:
- wer
model-index:
- name: whisper-small-italian-tuned
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Common Voice 4.0
type: mozilla-foundation/common_voice_4_0
config: it
split: test
args: 'config: it, split: test'
metrics:
- name: Wer
type: wer
value: 20.230225666742648
whisper-small-italian-tuned
This model is a fine-tuned version of openai/whisper-small on the Common Voice 4.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2653
- Wer: 20.2302
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
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: 300
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
0.2668 | 0.5647 | 1000 | 0.3023 | 22.8539 |
0.1314 | 1.1293 | 2000 | 0.2735 | 20.6998 |
0.118 | 1.6940 | 3000 | 0.2648 | 20.4263 |
0.0644 | 2.2586 | 4000 | 0.2653 | 20.2302 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.5.1+cu121
- Datasets 2.21.0
- Tokenizers 0.20.3