metadata
license: apache-2.0
tags:
- generated_from_trainer
base_model: paruwka/wav2vec2-base-timit-demo-google-colab
metrics:
- wer
model-index:
- name: wav2vec2-base-timit-demo-google-colab
results: []
wav2vec2-base-timit-demo-google-colab
This model is a fine-tuned version of paruwka/wav2vec2-base-timit-demo-google-colab on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0015
- Wer: 0.5828
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: 4
- 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
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
0.4127 | 4.46 | 500 | 0.8997 | 0.6221 |
0.38 | 8.93 | 1000 | 0.9162 | 0.6091 |
0.3382 | 13.39 | 1500 | 0.9044 | 0.6011 |
0.3074 | 17.86 | 2000 | 0.9529 | 0.6030 |
0.274 | 22.32 | 2500 | 0.9327 | 0.5919 |
0.2558 | 26.79 | 3000 | 0.9906 | 0.5895 |
0.2462 | 31.25 | 3500 | 0.9873 | 0.5854 |
0.2215 | 35.71 | 4000 | 1.0059 | 0.5839 |
0.2313 | 40.18 | 4500 | 1.0142 | 0.5823 |
0.2182 | 44.64 | 5000 | 0.9998 | 0.5804 |
0.2123 | 49.11 | 5500 | 1.0015 | 0.5828 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2