whisper-medium-nyagen-balanced-model
This model is a fine-tuned version of openai/whisper-medium on the nyagen dataset. It achieves the following results on the evaluation set:
- Loss: 0.3438
- Wer: 0.2336
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- 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 |
---|---|---|---|---|
1.8419 | 1.0756 | 200 | 0.5250 | 0.3745 |
0.6731 | 2.1511 | 400 | 0.3738 | 0.2701 |
0.3159 | 3.2267 | 600 | 0.3536 | 0.2448 |
0.2231 | 4.3023 | 800 | 0.3438 | 0.2336 |
0.1319 | 5.3779 | 1000 | 0.3518 | 0.2183 |
0.0553 | 6.4534 | 1200 | 0.3636 | 0.2954 |
0.062 | 7.5290 | 1400 | 0.3774 | 0.1984 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0
- Downloads last month
- 108
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Model tree for csikasote/whisper-medium-nyagen-balanced-model
Base model
openai/whisper-medium