<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
wav2vec2-base-common-voice-fa-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: 3.0558
- Wer: 1.0
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: 16
- 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: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
5.1626 | 0.3 | 100 | 4.0692 | 1.0 |
5.1776 | 0.6 | 200 | 3.6640 | 1.0 |
3.6628 | 0.9 | 300 | 3.3832 | 1.0 |
3.2022 | 1.2 | 400 | 3.3492 | 1.0 |
3.1714 | 1.5 | 500 | 3.3215 | 1.0 |
3.0689 | 1.8 | 600 | 3.0806 | 1.0 |
3.1478 | 2.1 | 700 | 3.0624 | 1.0 |
3.1818 | 2.4 | 800 | 3.0777 | 1.0 |
3.159 | 2.7 | 900 | 3.0558 | 1.0 |
Framework versions
- Transformers 4.11.3
- Pytorch 1.10.0+cu113
- Datasets 1.18.3
- Tokenizers 0.10.3