metadata
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- common_voice_13_0
metrics:
- wer
model-index:
- name: openai/whisper-small
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: common_voice_13_0
type: common_voice_13_0
config: pt
split: test
args: pt
metrics:
- name: Wer
type: wer
value: 10.414681431340979
openai/whisper-small
This model is a fine-tuned version of openai/whisper-small on the common_voice_13_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2528
- Wer: 10.4147
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: 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: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.1957 | 3.52 | 1000 | 0.2442 | 10.5938 |
| 0.1297 | 7.04 | 2000 | 0.2378 | 10.2849 |
| 0.0998 | 10.56 | 3000 | 0.2428 | 10.2520 |
| 0.0742 | 14.08 | 4000 | 0.2489 | 10.4015 |
| 0.0738 | 17.61 | 5000 | 0.2528 | 10.4147 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.0.1+cu117
- Datasets 2.14.4
- Tokenizers 0.15.1