metadata
library_name: peft
language:
- it
base_model: b-brave/asr_double_training_15-10-2024_merged
tags:
- generated_from_trainer
datasets:
- ASR_BB_and_EC
metrics:
- wer
model-index:
- name: Whisper Medium
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: ASR_BB_and_EC
type: ASR_BB_and_EC
config: default
split: test
args: default
metrics:
- type: wer
value: 36.18339529120198
name: Wer
Whisper Medium
This model is a fine-tuned version of b-brave/asr_double_training_15-10-2024_merged on the ASR_BB_and_EC dataset. It achieves the following results on the evaluation set:
- Loss: 0.4690
- Wer: 36.1834
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-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_steps: 50
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.9758 | 0.5464 | 50 | 0.4882 | 36.5551 |
| 0.81 | 1.0929 | 100 | 0.4839 | 36.5551 |
| 0.8005 | 1.6393 | 150 | 0.4808 | 36.4312 |
| 0.8837 | 2.1858 | 200 | 0.4782 | 35.8116 |
| 0.8172 | 2.7322 | 250 | 0.4760 | 35.8116 |
| 0.7142 | 3.2787 | 300 | 0.4742 | 35.9356 |
| 0.7817 | 3.8251 | 350 | 0.4724 | 35.9356 |
| 0.8025 | 4.3716 | 400 | 0.4707 | 36.0595 |
| 0.782 | 4.9180 | 450 | 0.4690 | 36.1834 |
Framework versions
- PEFT 0.13.2
- Transformers 4.45.2
- Pytorch 2.2.0
- Datasets 3.1.0
- Tokenizers 0.20.3