metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- PolyAI/minds14
metrics:
- wer
model-index:
- name: whisper-tiny-en-minds14
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: PolyAI/minds14
type: PolyAI/minds14
config: en-US
split: train
args: en-US
metrics:
- name: Wer
type: wer
value: 0.2744982290436836
whisper-tiny-en-minds14
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: 0.5680
- Wer Ortho: 0.2721
- Wer: 0.2745
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- 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
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 1.4576 | 1.79 | 50 | 0.9286 | 0.3128 | 0.3152 |
| 0.3694 | 3.57 | 100 | 0.5188 | 0.2776 | 0.2774 |
| 0.0466 | 5.36 | 150 | 0.4494 | 0.2640 | 0.2692 |
| 0.008 | 7.14 | 200 | 0.4855 | 0.2782 | 0.2816 |
| 0.0026 | 8.93 | 250 | 0.4892 | 0.2801 | 0.2845 |
| 0.0016 | 10.71 | 300 | 0.5116 | 0.2745 | 0.2774 |
| 0.0004 | 12.5 | 350 | 0.5383 | 0.2770 | 0.2798 |
| 0.0002 | 14.29 | 400 | 0.5471 | 0.2758 | 0.2774 |
| 0.0002 | 16.07 | 450 | 0.5590 | 0.2714 | 0.2733 |
| 0.0001 | 17.86 | 500 | 0.5680 | 0.2721 | 0.2745 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3