metadata
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- common_voice_13_0
metrics:
- wer
model-index:
- name: whisper_small_tw12
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: common_voice_13_0
type: common_voice_13_0
config: tw
split: train+test
args: tw
metrics:
- name: Wer
type: wer
value: 1.103734439834025
whisper_small_tw12
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: 3.3399
- Wer: 1.1037
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: 6.25e-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- training_steps: 500
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.0649 | 6.25 | 100 | 3.2733 | 1.5726 |
| 0.9932 | 12.5 | 200 | 2.9873 | 1.9378 |
| 0.0521 | 18.75 | 300 | 3.0893 | 1.1203 |
| 0.0045 | 25.0 | 400 | 3.2862 | 1.1245 |
| 0.0025 | 31.25 | 500 | 3.3399 | 1.1037 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.0