metadata
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- yt
metrics:
- wer
model-index:
- name: special2
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: yt id
type: yt
args: id
metrics:
- name: Wer
type: wer
value: 41.8877716509893
special2
This model is a fine-tuned version of openai/whisper-small on the yt id dataset. It achieves the following results on the evaluation set:
- Loss: 0.6621
- Wer: 41.8878
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: 12
- eval_batch_size: 6
- 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
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.8117 | 0.26 | 1000 | 0.8215 | 48.6085 |
| 0.7087 | 0.52 | 2000 | 0.7323 | 52.3062 |
| 0.7057 | 0.77 | 3000 | 0.6922 | 50.0032 |
| 0.5319 | 1.03 | 4000 | 0.6686 | 42.1213 |
| 0.4704 | 1.29 | 5000 | 0.6621 | 41.8878 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3