wav2vec
This model is a fine-tuned version of facebook/wav2vec2-base on the speech_commands dataset. It achieves the following results on the evaluation set:
- Loss: 0.4992
- Accuracy: 0.8939
- F1: 0.8872
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: 3e-05
- train_batch_size: 80
- eval_batch_size: 80
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.6895 | 1.0 | 639 | 0.7875 | 0.8773 | 0.7995 |
| 0.4171 | 2.0 | 1278 | 0.5445 | 0.8932 | 0.8675 |
| 0.2706 | 3.0 | 1917 | 0.4992 | 0.8939 | 0.8872 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0
- Downloads last month
- 4
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.
Model tree for Zarakun/wav2vec
Base model
facebook/wav2vec2-baseDataset used to train Zarakun/wav2vec
Evaluation results
- Accuracy on speech_commandstest set self-reported0.894
- F1 on speech_commandstest set self-reported0.887