metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
datasets:
- vivos
metrics:
- wer
model-index:
- name: wav2vec2-vivos-asr
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: vivos
type: vivos
config: default
split: None
args: default
metrics:
- name: Wer
type: wer
value: 0.20053544110147883
wav2vec2-vivos-asr
This model is a fine-tuned version of facebook/wav2vec2-base on the vivos dataset. It achieves the following results on the evaluation set:
- Loss: 0.3151
- Wer: 0.2005
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: 32
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 300
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 7.4472 | 2.0 | 146 | 4.1025 | 1.0 |
| 3.4729 | 4.0 | 292 | 3.6046 | 1.0 |
| 2.8216 | 6.0 | 438 | 1.4275 | 0.8461 |
| 0.8343 | 8.0 | 584 | 0.5637 | 0.3727 |
| 0.4258 | 10.0 | 730 | 0.4272 | 0.2877 |
| 0.3012 | 12.0 | 876 | 0.3548 | 0.2370 |
| 0.2445 | 14.0 | 1022 | 0.3371 | 0.2230 |
| 0.2328 | 16.0 | 1168 | 0.3269 | 0.2119 |
| 0.1831 | 18.0 | 1314 | 0.3204 | 0.2058 |
| 0.1664 | 20.0 | 1460 | 0.3151 | 0.2005 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1