metadata
license: mit
base_model: facebook/w2v-bert-2.0
tags:
- generated_from_trainer
datasets:
- common_voice_16_0
metrics:
- wer
model-index:
- name: w2v-bertkmr-test
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: common_voice_16_0
type: common_voice_16_0
config: kmr
split: test
args: kmr
metrics:
- name: Wer
type: wer
value: 0.1570856537948175
w2v-bertkmr-test
This model is a fine-tuned version of facebook/w2v-bert-2.0 on the common_voice_16_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2399
- Wer: 0.1571
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 150
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| No log | 0.8 | 200 | 0.3476 | 0.3257 |
| 1.2561 | 1.6 | 400 | 0.2756 | 0.2669 |
| 0.1906 | 2.4 | 600 | 0.2484 | 0.2363 |
| 0.1906 | 3.2 | 800 | 0.2336 | 0.2177 |
| 0.1242 | 4.0 | 1000 | 0.2192 | 0.1919 |
| 0.0853 | 4.8 | 1200 | 0.2217 | 0.1879 |
| 0.0853 | 5.6 | 1400 | 0.2272 | 0.1786 |
| 0.0586 | 6.4 | 1600 | 0.2292 | 0.1695 |
| 0.0365 | 7.2 | 1800 | 0.2276 | 0.1613 |
| 0.0365 | 8.0 | 2000 | 0.2127 | 0.1626 |
| 0.0222 | 8.8 | 2200 | 0.2271 | 0.1568 |
| 0.0118 | 9.6 | 2400 | 0.2399 | 0.1571 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1