---
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: xls-r-1b-bemba-15hrs
  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. -->

# xls-r-1b-bemba-15hrs

This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2134
- Wer: 0.3485

## 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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- 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: 400
- num_epochs: 10
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 2.3016        | 0.36  | 400  | 0.6032          | 0.9932 |
| 0.5196        | 0.71  | 800  | 0.3089          | 0.5020 |
| 0.4397        | 1.07  | 1200 | 0.2562          | 0.4223 |
| 0.3617        | 1.43  | 1600 | 0.2269          | 0.4009 |
| 0.36          | 1.79  | 2000 | 0.2106          | 0.3896 |
| 0.3404        | 2.14  | 2400 | 0.2079          | 0.3681 |
| 0.2915        | 2.5   | 2800 | 0.2024          | 0.3488 |
| 0.2869        | 2.86  | 3200 | 0.2068          | 0.3550 |
| 0.2492        | 3.22  | 3600 | 0.1925          | 0.3273 |
| 0.2542        | 3.57  | 4000 | 0.2041          | 0.3446 |
| 0.2333        | 3.93  | 4400 | 0.1985          | 0.3386 |
| 0.2023        | 4.29  | 4800 | 0.2134          | 0.3485 |


### Framework versions

- Transformers 4.19.0.dev0
- Pytorch 1.10.0+cu111
- Datasets 2.1.0
- Tokenizers 0.12.1