metadata
license: apache-2.0
tags:
- generated_from_trainer
base_model: facebook/wav2vec2-large-xlsr-53
datasets:
- common_voice_13_0
metrics:
- wer
model-index:
- name: wav2vec2-xlsr-53-CV-demo-google-colab-Ezra_William_Prod13
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: 0.4928097345132743
name: Wer
wav2vec2-xlsr-53-CV-demo-google-colab-Ezra_William_Prod13
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the common_voice_13_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4972
- Wer: 0.4928
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 9
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 4.7378 | 0.9 | 500 | 2.9498 | 1.0 |
| 2.91 | 1.8 | 1000 | 2.8716 | 1.0 |
| 2.683 | 2.7 | 1500 | 1.9348 | 1.0 |
| 1.5179 | 3.6 | 2000 | 0.8042 | 0.6992 |
| 1.014 | 4.5 | 2500 | 0.6370 | 0.5932 |
| 0.87 | 5.4 | 3000 | 0.5648 | 0.5443 |
| 0.795 | 6.29 | 3500 | 0.5328 | 0.5177 |
| 0.742 | 7.19 | 4000 | 0.5148 | 0.5016 |
| 0.701 | 8.09 | 4500 | 0.4969 | 0.4943 |
| 0.7002 | 8.99 | 5000 | 0.4972 | 0.4928 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2