Whisper Large v3 cmb
This model is a fine-tuned version of openai/whisper-large-v3 on the Common Voice 13, Google Fleurs and juzne vesti dataset. It achieves the following results on the evaluation set:
- Loss: 0.1111
- Wer Ortho: 0.1339
- Wer: 0.0415
Model description
Dataset Juzne vesti is published by
Rupnik, Peter and Ljubešić, Nikola, 2022,
ASR training dataset for Serbian JuzneVesti-SR v1.0, Slovenian language resource repository CLARIN.SI, ISSN 2820-4042,
http://hdl.handle.net/11356/1679.
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: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 1500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.2766 | 0.48 | 500 | 0.1350 | 0.1670 | 0.0595 |
| 0.2813 | 0.95 | 1000 | 0.1134 | 0.1426 | 0.0491 |
| 0.1858 | 1.43 | 1500 | 0.1111 | 0.1339 | 0.0415 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.14.1