| --- |
| library_name: transformers |
| license: apache-2.0 |
| base_model: openai/whisper-tiny |
| tags: |
| - generated_from_trainer |
| datasets: |
| - PolyAI/minds14 |
| metrics: |
| - wer |
| model-index: |
| - name: my-seq2seq-model |
| results: |
| - task: |
| name: Automatic Speech Recognition |
| type: automatic-speech-recognition |
| dataset: |
| name: PolyAI/minds14 |
| type: PolyAI/minds14 |
| metrics: |
| - name: Wer |
| type: wer |
| value: 0.35424354243542433 |
| --- |
| |
| <!-- This model card has been generated automatically according to the information the Trainer had access to. You |
| should probably proofread and complete it, then remove this comment. --> |
|
|
| # my-seq2seq-model |
|
|
| This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.7355 |
| - Wer Ortho: 0.3526 |
| - Wer: 0.3542 |
|
|
| ## 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: 5e-05 |
| - train_batch_size: 8 |
| - eval_batch_size: 8 |
| - seed: 42 |
| - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
| - lr_scheduler_type: linear |
| - num_epochs: 40 |
| - mixed_precision_training: Native AMP |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer | |
| |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:| |
| | No log | 1.0 | 57 | 0.5982 | 0.3546 | 0.3536 | |
| | No log | 2.0 | 114 | 0.6244 | 0.3816 | 0.3838 | |
| | No log | 3.0 | 171 | 0.6240 | 0.3662 | 0.3629 | |
| | No log | 4.0 | 228 | 0.6321 | 0.3423 | 0.3426 | |
| | No log | 5.0 | 285 | 0.6522 | 0.3732 | 0.3715 | |
| | No log | 6.0 | 342 | 0.6582 | 0.3526 | 0.3524 | |
| | No log | 7.0 | 399 | 0.6628 | 0.3507 | 0.3506 | |
| | No log | 8.0 | 456 | 0.6683 | 0.3501 | 0.3499 | |
| | 0.0203 | 9.0 | 513 | 0.6734 | 0.3443 | 0.3444 | |
| | 0.0203 | 10.0 | 570 | 0.6804 | 0.3456 | 0.3456 | |
| | 0.0203 | 11.0 | 627 | 0.6837 | 0.3430 | 0.3432 | |
| | 0.0203 | 12.0 | 684 | 0.6878 | 0.3468 | 0.3456 | |
| | 0.0203 | 13.0 | 741 | 0.6939 | 0.3481 | 0.3469 | |
| | 0.0203 | 14.0 | 798 | 0.6960 | 0.3501 | 0.3487 | |
| | 0.0203 | 15.0 | 855 | 0.6991 | 0.3481 | 0.3469 | |
| | 0.0203 | 16.0 | 912 | 0.7017 | 0.3468 | 0.3475 | |
| | 0.0203 | 17.0 | 969 | 0.7044 | 0.3468 | 0.3475 | |
| | 0.0024 | 18.0 | 1026 | 0.7082 | 0.3507 | 0.3506 | |
| | 0.0024 | 19.0 | 1083 | 0.7089 | 0.3507 | 0.3506 | |
| | 0.0024 | 20.0 | 1140 | 0.7131 | 0.3533 | 0.3536 | |
| | 0.0024 | 21.0 | 1197 | 0.7141 | 0.3546 | 0.3549 | |
| | 0.0024 | 22.0 | 1254 | 0.7176 | 0.3520 | 0.3518 | |
| | 0.0024 | 23.0 | 1311 | 0.7197 | 0.3507 | 0.3506 | |
| | 0.0024 | 24.0 | 1368 | 0.7211 | 0.3507 | 0.3506 | |
| | 0.0024 | 25.0 | 1425 | 0.7223 | 0.3501 | 0.3499 | |
| | 0.0024 | 26.0 | 1482 | 0.7243 | 0.3501 | 0.3499 | |
| | 0.0014 | 27.0 | 1539 | 0.7255 | 0.3507 | 0.3506 | |
| | 0.0014 | 28.0 | 1596 | 0.7264 | 0.3514 | 0.3512 | |
| | 0.0014 | 29.0 | 1653 | 0.7293 | 0.3520 | 0.3524 | |
| | 0.0014 | 30.0 | 1710 | 0.7297 | 0.3514 | 0.3518 | |
| | 0.0014 | 31.0 | 1767 | 0.7310 | 0.3507 | 0.3518 | |
| | 0.0014 | 32.0 | 1824 | 0.7308 | 0.3507 | 0.3518 | |
| | 0.0014 | 33.0 | 1881 | 0.7330 | 0.3507 | 0.3518 | |
| | 0.0014 | 34.0 | 1938 | 0.7336 | 0.3507 | 0.3518 | |
| | 0.0014 | 35.0 | 1995 | 0.7338 | 0.3507 | 0.3518 | |
| | 0.0010 | 36.0 | 2052 | 0.7341 | 0.3526 | 0.3542 | |
| | 0.0010 | 37.0 | 2109 | 0.7349 | 0.3520 | 0.3530 | |
| | 0.0010 | 38.0 | 2166 | 0.7350 | 0.3526 | 0.3542 | |
| | 0.0010 | 39.0 | 2223 | 0.7356 | 0.3520 | 0.3530 | |
| | 0.0010 | 40.0 | 2280 | 0.7355 | 0.3526 | 0.3542 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 5.14.1 |
| - Pytorch 2.11.0+cu128 |
| - Datasets 5.0.0 |
| - Tokenizers 0.22.2 |
| |
