---
license: mit
tags:
- generated_from_trainer
base_model: facebook/w2v-bert-2.0
datasets:
- generator
metrics:
- wer
model-index:
- name: wav2vec2-bert-fon
  results:
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset:
      name: generator
      type: generator
      config: default
      split: train
      args: default
    metrics:
    - type: wer
      value: 0.2183013319494897
      name: Wer
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# wav2vec2-bert-fon

This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on the generator dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2480
- Wer: 0.2183

## 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: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 8
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Wer    |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 1.7256        | 0.35  | 500   | 0.7130          | 0.5735 |
| 0.497         | 0.7   | 1000  | 0.4485          | 0.4113 |
| 0.4133        | 1.05  | 1500  | 0.3067          | 0.2729 |
| 0.303         | 1.4   | 2000  | 0.2747          | 0.2340 |
| 0.2556        | 1.75  | 2500  | 0.2047          | 0.1793 |
| 0.2378        | 2.1   | 3000  | 0.1794          | 0.1528 |
| 0.1619        | 2.45  | 3500  | 0.1831          | 0.1526 |
| 0.1702        | 2.8   | 4000  | 0.1707          | 0.1275 |
| 0.142         | 3.15  | 4500  | 0.1653          | 0.1273 |
| 0.133         | 3.5   | 5000  | 0.1467          | 0.1218 |
| 0.1864        | 3.85  | 5500  | 0.2228          | 0.1565 |
| 0.3031        | 4.21  | 6000  | 0.2462          | 0.2076 |
| 0.3263        | 4.56  | 6500  | 0.2432          | 0.1713 |
| 0.3354        | 4.91  | 7000  | 0.2506          | 0.2277 |
| 0.3266        | 5.26  | 7500  | 0.2480          | 0.2183 |
| 0.3247        | 5.61  | 8000  | 0.2480          | 0.2183 |
| 0.3416        | 5.96  | 8500  | 0.2480          | 0.2183 |
| 0.323         | 6.31  | 9000  | 0.2480          | 0.2183 |
| 0.3303        | 6.66  | 9500  | 0.2480          | 0.2183 |
| 0.3328        | 7.01  | 10000 | 0.2480          | 0.2183 |
| 0.3244        | 7.36  | 10500 | 0.2480          | 0.2183 |
| 0.3362        | 7.71  | 11000 | 0.2480          | 0.2183 |


### Framework versions

- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2