---
license: cc-by-nc-4.0
base_model: facebook/mms-1b-all
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2-large-mms-1b-tira-lid
  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. -->

# wav2vec2-large-mms-1b-tira-lid

This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0473
- Wer: 0.0167

## 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: 0.001
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 2
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 4

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 0.2876        | 0.42  | 100  | 0.0493          | 0.0167 |
| 0.2922        | 0.84  | 200  | 0.1215          | 0.05   |
| 0.2377        | 1.26  | 300  | 0.0741          | 0.0333 |
| 0.1442        | 1.67  | 400  | 0.1658          | 0.0333 |
| 0.0873        | 2.09  | 500  | 0.0204          | 0.0    |
| 0.0695        | 2.51  | 600  | 0.0691          | 0.0167 |
| 0.075         | 2.93  | 700  | 0.0758          | 0.0333 |
| 0.0412        | 3.35  | 800  | 0.0650          | 0.0167 |
| 0.0523        | 3.77  | 900  | 0.0473          | 0.0167 |


### Framework versions

- Transformers 4.32.1
- Pytorch 2.0.1+cu117
- Datasets 2.14.4
- Tokenizers 0.13.3