metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- fleurs
metrics:
- wer
model-index:
- name: openai/whisper-tiny
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: fleurs
type: fleurs
config: sv_se
split: validation
args: sv_se
metrics:
- name: Wer
type: wer
value: 168.6092926712438
openai/whisper-tiny
This model is a fine-tuned version of openai/whisper-tiny on the fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 1.0456
- Wer: 168.6093
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: 64
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- training_steps: 112
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 1.5299 | 0.1 | 11 | 1.5622 | 219.6711 |
| 1.1908 | 0.2 | 22 | 1.3652 | 192.2401 |
| 1.1161 | 0.29 | 33 | 1.1921 | 200.2395 |
| 0.9216 | 1.05 | 44 | 1.1263 | 186.5240 |
| 0.8441 | 1.15 | 55 | 1.0946 | 179.3230 |
| 0.8505 | 1.25 | 66 | 1.0748 | 159.6839 |
| 0.7844 | 2.01 | 77 | 1.0585 | 163.2924 |
| 0.7208 | 2.11 | 88 | 1.0491 | 158.1031 |
| 0.6481 | 2.21 | 99 | 1.0468 | 158.5183 |
| 0.7912 | 2.3 | 110 | 1.0456 | 168.6093 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2