---
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: mr_sam_wav2vec2_nigerian_accent_v1
  results: []
---

<!-- 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. -->

# mr_sam_wav2vec2_nigerian_accent_v1

This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3400
- Wer: 0.1900

## 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
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 40
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 4.456         | 6.67  | 500  | 2.9391          | 1.0    |
| 1.0329        | 13.33 | 1000 | 0.4350          | 0.3343 |
| 0.2048        | 20.0  | 1500 | 0.3604          | 0.2488 |
| 0.1054        | 26.67 | 2000 | 0.3545          | 0.2118 |
| 0.0698        | 33.33 | 2500 | 0.3480          | 0.2012 |
| 0.0529        | 40.0  | 3000 | 0.3400          | 0.1900 |


### Framework versions

- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2