metadata
library_name: transformers
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
datasets:
- PolyAI/minds14
metrics:
- wer
model-index:
- name: whisper-tiny-en-US
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: PolyAI/minds14
type: PolyAI/minds14
config: en-US
split: train[450:]
args: en-US
metrics:
- name: Wer
type: wer
value: 0.34887839433293977
whisper-tiny-en-US
This model is a fine-tuned version of openai/whisper-tiny on the PolyAI/minds14 dataset. It achieves the following results on the evaluation set:
- Loss: 0.6638
- Wer Ortho: 34.5466
- Wer: 0.3489
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: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 50
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.7657 | 1.7857 | 50 | 0.5870 | 39.4818 | 0.3932 |
| 0.2562 | 3.5714 | 100 | 0.4866 | 34.8550 | 0.3483 |
| 0.0666 | 5.3571 | 150 | 0.5190 | 34.5466 | 0.3489 |
| 0.0228 | 7.1429 | 200 | 0.5649 | 32.4491 | 0.3288 |
| 0.0065 | 8.9286 | 250 | 0.5845 | 32.0173 | 0.3229 |
| 0.0018 | 10.7143 | 300 | 0.6142 | 33.6212 | 0.3400 |
| 0.0012 | 12.5 | 350 | 0.6320 | 33.3128 | 0.3371 |
| 0.0008 | 14.2857 | 400 | 0.6443 | 34.1764 | 0.3465 |
| 0.0007 | 16.0714 | 450 | 0.6548 | 34.2381 | 0.3447 |
| 0.0007 | 17.8571 | 500 | 0.6638 | 34.5466 | 0.3489 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1