Whisper tiny Ta example - Bharat Ramanathan
This model is a fine-tuned version of parambharat/whisper-tiny-ta on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4016
- Wer: 36.5217
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: 32
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 25
- training_steps: 100
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
0.304 | 12.01 | 25 | 0.3614 | 31.7391 |
0.1826 | 24.02 | 50 | 0.3851 | 35.2174 |
0.1346 | 37.01 | 75 | 0.3999 | 37.8261 |
0.1096 | 49.02 | 100 | 0.4016 | 36.5217 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2
- Downloads last month
- 6
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.