metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- fleurs
metrics:
- wer
model-index:
- name: openai/whisper-small
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: fleurs
type: fleurs
config: cmn_hans_cn
split: test
args: cmn_hans_cn
metrics:
- name: Wer
type: wer
value: 16.05478120919498
openai/whisper-small
This model is a fine-tuned version of openai/whisper-small on the fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 0.3537
- Wer: 16.0548
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: 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 |
|---|---|---|---|---|
| 0.0013 | 19.01 | 1000 | 0.2910 | 14.9282 |
| 0.0005 | 39.0 | 2000 | 0.3185 | 15.6327 |
| 0.0002 | 58.01 | 3000 | 0.3369 | 15.6965 |
| 0.0002 | 78.0 | 4000 | 0.3480 | 16.0669 |
| 0.0001 | 98.0 | 5000 | 0.3537 | 16.0548 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2