metadata
license: apache-2.0
tags:
- generated_from_trainer
base_model: facebook/wav2vec2-xls-r-300m
datasets:
- common_voice_13_0
metrics:
- wer
model-index:
- name: wav2vec2-xlsr-53-CV-demo-google-colab-Ezra_William_Prod16
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: common_voice_13_0
type: common_voice_13_0
config: id
split: test
args: id
metrics:
- type: wer
value: 0.31356010324483774
name: Wer
wav2vec2-xlsr-53-CV-demo-google-colab-Ezra_William_Prod16
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice_13_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3425
- Wer: 0.3136
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
- num_epochs: 12
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.9562 | 1.0 | 278 | 2.9246 | 1.0 |
| 2.8689 | 2.0 | 556 | 2.7856 | 1.0 |
| 1.1264 | 3.0 | 834 | 0.6656 | 0.6370 |
| 0.6913 | 4.0 | 1112 | 0.5009 | 0.5000 |
| 0.5835 | 5.0 | 1390 | 0.4370 | 0.4477 |
| 0.4684 | 6.0 | 1668 | 0.3923 | 0.3857 |
| 0.4139 | 7.0 | 1946 | 0.3876 | 0.3761 |
| 0.389 | 8.0 | 2224 | 0.3881 | 0.3551 |
| 0.3559 | 9.0 | 2502 | 0.3580 | 0.3387 |
| 0.3228 | 10.0 | 2780 | 0.3552 | 0.3249 |
| 0.3295 | 11.0 | 3058 | 0.3385 | 0.3197 |
| 0.3112 | 12.0 | 3336 | 0.3425 | 0.3136 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.2.2+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1