960h
This model is a fine-tuned version of facebook/wav2vec2-base-960h on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5325
- Wer: 24.4442
- Cer: 12.0113
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: 0.0001
- train_batch_size: 32
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2000
- training_steps: 2700
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 1.0759 | 1.75 | 500 | 0.7781 | 32.9761 | 16.6493 |
| 0.8049 | 3.51 | 1000 | 0.6552 | 29.8584 | 14.9333 |
| 0.7047 | 5.26 | 1500 | 0.6050 | 27.5232 | 13.6151 |
| 0.6339 | 7.02 | 2000 | 0.5865 | 26.3889 | 13.1279 |
| 0.5331 | 8.77 | 2500 | 0.5325 | 24.4442 | 12.0113 |
Framework versions
- Transformers 4.39.3
- Pytorch 1.12.1+cu116
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 160
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 ImanNalia/ft_wav2vec2_960h_thirty
Base model
facebook/wav2vec2-base-960h