openai/whisper-large-v2
This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1915
- Wer: 10.0336
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: 32
- 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: 4000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
0.2819 | 0.12 | 500 | 0.2709 | 15.1530 |
0.2508 | 0.25 | 1000 | 0.2098 | 11.2876 |
0.1113 | 1.01 | 1500 | 0.2127 | 10.3778 |
0.2872 | 1.14 | 2000 | 0.1891 | 10.6509 |
0.2995 | 1.26 | 2500 | 0.1883 | 10.7545 |
0.0701 | 2.02 | 3000 | 0.1972 | 9.6061 |
0.0613 | 2.15 | 3500 | 0.2073 | 9.4813 |
0.1135 | 2.27 | 4000 | 0.1915 | 10.0336 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.9.1.dev0
- Tokenizers 0.13.2
- Downloads last month
- 0
Evaluation results
- WER on rishabhjain16/infer_mysttest set self-reported11.730
- WER on rishabhjain16/infer_pfstest set self-reported3.130
- WER on rishabhjain16/infer_cmutest set self-reported2.560
- WER on rishabhjain16/libritts_dev_cleantest set self-reported4.690
- WER on rishabhjain16/infer_pf_swedishtest set self-reported9.670
- WER on rishabhjain16/infer_pf_germantest set self-reported35.050
- WER on rishabhjain16/infer_pf_italiantest set self-reported5.510
- WER on rishabhjain16/infer_so_chinesetest set self-reported15.830