Whisper medium-translate Hi - Aa
This model is a fine-tuned version of Aakali/whisper-medium-hi on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:
- Loss: 2.4968
- Wer: 23.6842
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: 8
- 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: 5000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
0.0 | 1000.0 | 1000 | 1.2182 | 13.1579 |
0.0 | 2000.0 | 2000 | 1.7360 | 18.4211 |
0.0 | 3000.0 | 3000 | 2.1484 | 23.6842 |
0.0 | 4000.0 | 4000 | 2.5106 | 26.3158 |
0.0 | 5000.0 | 5000 | 2.4968 | 23.6842 |
Framework versions
- Transformers 4.41.0
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 1
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.