metadata
language:
- en
license: apache-2.0
base_model: openai/whisper-small
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- LanugaugeLab
metrics:
- wer
model-index:
- name: Whisper Small - LanguageLab V1
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Language P1
type: LanugaugeLab
config: default
split: train
args: 'config: hi, split: test'
metrics:
- name: Wer
type: wer
value: 31.753554502369667
Whisper Small - LanguageLab V1
This model is a fine-tuned version of openai/whisper-small on the Language P1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1019
- Wer: 31.7536
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: 16
- eval_batch_size: 8
- 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 |
|---|---|---|---|---|
| 0.0001 | 21.28 | 1000 | 0.0898 | 43.6019 |
| 0.0 | 42.55 | 2000 | 0.0958 | 36.9668 |
| 0.0 | 63.83 | 3000 | 0.0991 | 35.0711 |
| 0.0 | 85.11 | 4000 | 0.1011 | 32.7014 |
| 0.0 | 106.38 | 5000 | 0.1019 | 31.7536 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu118
- Datasets 2.17.1
- Tokenizers 0.15.2