wav2vec2-base-timit-demo-colab
This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4374
- Wer: 0.3392
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
3.364 | 4.0 | 500 | 1.2339 | 0.9609 |
0.5751 | 8.0 | 1000 | 0.4374 | 0.4287 |
0.2175 | 12.0 | 1500 | 0.4184 | 0.3809 |
0.1281 | 16.0 | 2000 | 0.4357 | 0.3720 |
0.0889 | 20.0 | 2500 | 0.4152 | 0.3481 |
0.0604 | 24.0 | 3000 | 0.4518 | 0.3447 |
0.0471 | 28.0 | 3500 | 0.4374 | 0.3392 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 1.18.3
- Tokenizers 0.19.1
- Downloads last month
- 5
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.