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_Prod12
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: 0.29217367256637167
name: Wer
wav2vec2-xlsr-53-CV-demo-google-colab-Ezra_William_Prod12
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: 0.3025
- Wer: 0.2922
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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.9131 | 1.0 | 464 | 2.9568 | 1.0 |
| 1.0005 | 2.0 | 928 | 0.6123 | 0.5699 |
| 0.5939 | 3.0 | 1392 | 0.4082 | 0.4204 |
| 0.4569 | 4.0 | 1856 | 0.3475 | 0.3725 |
| 0.4151 | 5.0 | 2320 | 0.3333 | 0.3413 |
| 0.3655 | 6.0 | 2784 | 0.3223 | 0.3234 |
| 0.3351 | 7.0 | 3248 | 0.3163 | 0.3078 |
| 0.3103 | 8.0 | 3712 | 0.3045 | 0.2956 |
| 0.3063 | 9.0 | 4176 | 0.3001 | 0.2900 |
| 0.3072 | 10.0 | 4640 | 0.3025 | 0.2922 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2