metadata
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
datasets:
- PolyAI/minds14
metrics:
- wer
model-index:
- name: whisper-tiny-enUS
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: 7.431944109853047
whisper-tiny-enUS
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: 0.0000
- Wer Ortho: 7.4972
- Wer: 7.4319
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: 4000
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.0017 | 14.29 | 500 | 0.0011 | 4.3200 | 4.2159 |
| 0.0003 | 28.57 | 1000 | 0.0005 | 4.4204 | 4.3724 |
| 0.001 | 42.86 | 1500 | 0.0003 | 4.1567 | 4.0954 |
| 0.0001 | 57.14 | 2000 | 0.0001 | 4.3702 | 4.3483 |
| 0.0001 | 71.43 | 2500 | 0.0001 | 7.1958 | 7.1429 |
| 0.0 | 85.71 | 3000 | 0.0000 | 7.5097 | 7.4440 |
| 0.0 | 100.0 | 3500 | 0.0000 | 7.5348 | 7.4681 |
| 0.0 | 114.29 | 4000 | 0.0000 | 7.4972 | 7.4319 |
Framework versions
- Transformers 4.32.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3