---
library_name: transformers
license: cc-by-nc-4.0
base_model: facebook/mms-1b-all
tags:
- generated_from_trainer
datasets:
- xtreme_s
metrics:
- wer
model-index:
- name: wav2vec2-large-mms-1b-somalia
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: xtreme_s
      type: xtreme_s
      config: fleurs.so_so
      split: test
      args: fleurs.so_so
    metrics:
    - name: Wer
      type: wer
      value: 0.4330026371970363
---

<!-- 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-somalia

This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the xtreme_s dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5553
- Wer: 0.4330

## 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: 2
- eval_batch_size: 8
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 4
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Wer    |
|:-------------:|:------:|:----:|:---------------:|:------:|
| 6.5097        | 0.0558 | 100  | 0.6980          | 0.4802 |
| 0.4771        | 0.1117 | 200  | 0.6315          | 0.4613 |
| 0.4098        | 0.1675 | 300  | 0.6112          | 0.4519 |
| 0.4555        | 0.2233 | 400  | 0.6085          | 0.4540 |
| 0.4319        | 0.2792 | 500  | 0.6041          | 0.4614 |
| 0.4582        | 0.3350 | 600  | 0.5927          | 0.4526 |
| 0.3879        | 0.3908 | 700  | 0.6079          | 0.4530 |
| 0.3919        | 0.4467 | 800  | 0.6135          | 0.4479 |
| 0.4245        | 0.5025 | 900  | 0.5879          | 0.4461 |
| 0.3872        | 0.5583 | 1000 | 0.6088          | 0.4627 |
| 0.3843        | 0.6142 | 1100 | 0.5939          | 0.4451 |
| 0.3842        | 0.6700 | 1200 | 0.6133          | 0.4631 |
| 0.3771        | 0.7259 | 1300 | 0.6026          | 0.4510 |
| 0.4458        | 0.7817 | 1400 | 0.5965          | 0.4494 |
| 0.4114        | 0.8375 | 1500 | 0.5950          | 0.4472 |
| 0.4135        | 0.8934 | 1600 | 0.6106          | 0.4613 |
| 0.3763        | 0.9492 | 1700 | 0.6082          | 0.4426 |
| 0.442         | 1.0050 | 1800 | 0.5922          | 0.4482 |
| 0.381         | 1.0609 | 1900 | 0.5827          | 0.4419 |
| 0.3874        | 1.1167 | 2000 | 0.5816          | 0.4518 |
| 0.4602        | 1.1725 | 2100 | 0.5888          | 0.4453 |
| 0.3709        | 1.2284 | 2200 | 0.5870          | 0.4447 |
| 0.3991        | 1.2842 | 2300 | 0.5749          | 0.4426 |
| 0.3771        | 1.3400 | 2400 | 0.5743          | 0.4396 |
| 0.351         | 1.3959 | 2500 | 0.5956          | 0.4414 |
| 0.358         | 1.4517 | 2600 | 0.5772          | 0.4368 |
| 0.4076        | 1.5075 | 2700 | 0.5931          | 0.4384 |
| 0.4322        | 1.5634 | 2800 | 0.5795          | 0.4407 |
| 0.359         | 1.6192 | 2900 | 0.5814          | 0.4408 |
| 0.3714        | 1.6750 | 3000 | 0.5794          | 0.4366 |
| 0.407         | 1.7309 | 3100 | 0.5673          | 0.4404 |
| 0.3803        | 1.7867 | 3200 | 0.5754          | 0.4453 |
| 0.3412        | 1.8425 | 3300 | 0.5974          | 0.4470 |
| 0.3826        | 1.8984 | 3400 | 0.5847          | 0.4422 |
| 0.3961        | 1.9542 | 3500 | 0.5800          | 0.4417 |
| 0.4035        | 2.0101 | 3600 | 0.5804          | 0.4395 |
| 0.3988        | 2.0659 | 3700 | 0.5626          | 0.4379 |
| 0.3642        | 2.1217 | 3800 | 0.5745          | 0.4393 |
| 0.3543        | 2.1776 | 3900 | 0.5786          | 0.4414 |
| 0.3577        | 2.2334 | 4000 | 0.5736          | 0.4376 |
| 0.3818        | 2.2892 | 4100 | 0.5745          | 0.4412 |
| 0.3538        | 2.3451 | 4200 | 0.5856          | 0.4370 |
| 0.3902        | 2.4009 | 4300 | 0.5708          | 0.4370 |
| 0.3915        | 2.4567 | 4400 | 0.5819          | 0.4372 |
| 0.3918        | 2.5126 | 4500 | 0.5620          | 0.4366 |
| 0.3716        | 2.5684 | 4600 | 0.5801          | 0.4497 |
| 0.3696        | 2.6242 | 4700 | 0.5640          | 0.4321 |
| 0.3532        | 2.6801 | 4800 | 0.5772          | 0.4418 |
| 0.4039        | 2.7359 | 4900 | 0.5660          | 0.4388 |
| 0.3488        | 2.7917 | 5000 | 0.5666          | 0.4364 |
| 0.3482        | 2.8476 | 5100 | 0.5662          | 0.4428 |
| 0.3793        | 2.9034 | 5200 | 0.5620          | 0.4368 |
| 0.3808        | 2.9592 | 5300 | 0.5559          | 0.4387 |
| 0.3403        | 3.0151 | 5400 | 0.5647          | 0.4387 |
| 0.3654        | 3.0709 | 5500 | 0.5515          | 0.4367 |
| 0.3614        | 3.1267 | 5600 | 0.5593          | 0.4410 |
| 0.3399        | 3.1826 | 5700 | 0.5604          | 0.4353 |
| 0.3627        | 3.2384 | 5800 | 0.5559          | 0.4379 |
| 0.3583        | 3.2942 | 5900 | 0.5564          | 0.4343 |
| 0.3897        | 3.3501 | 6000 | 0.5593          | 0.4362 |
| 0.3523        | 3.4059 | 6100 | 0.5578          | 0.4345 |
| 0.3412        | 3.4618 | 6200 | 0.5586          | 0.4346 |
| 0.3877        | 3.5176 | 6300 | 0.5529          | 0.4346 |
| 0.3605        | 3.5734 | 6400 | 0.5577          | 0.4346 |
| 0.3869        | 3.6293 | 6500 | 0.5522          | 0.4332 |
| 0.3906        | 3.6851 | 6600 | 0.5591          | 0.4312 |
| 0.3469        | 3.7409 | 6700 | 0.5585          | 0.4351 |
| 0.3665        | 3.7968 | 6800 | 0.5554          | 0.4333 |
| 0.3851        | 3.8526 | 6900 | 0.5543          | 0.4346 |
| 0.3195        | 3.9084 | 7000 | 0.5558          | 0.4348 |
| 0.346         | 3.9643 | 7100 | 0.5553          | 0.4330 |


### Framework versions

- Transformers 4.48.0.dev0
- Pytorch 2.1.0+cu118
- Datasets 3.2.0
- Tokenizers 0.21.0