---
library_name: transformers
license: cc-by-nc-4.0
base_model: facebook/mms-1b-all
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: mms_eng_yor
  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. -->

# mms_eng_yor

This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6192
- Wer: 0.5316

## 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: 5e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Wer    |
|:-------------:|:------:|:----:|:---------------:|:------:|
| 10.7343       | 0.2436 | 100  | 4.3854          | 1.0    |
| 3.1325        | 0.4872 | 200  | 2.1582          | 0.9691 |
| 1.6166        | 0.7308 | 300  | 1.1347          | 0.7188 |
| 1.1808        | 0.9744 | 400  | 0.9792          | 0.6747 |
| 1.0459        | 1.2168 | 500  | 0.9050          | 0.6494 |
| 1.0317        | 1.4604 | 600  | 0.8543          | 0.6338 |
| 0.9836        | 1.7040 | 700  | 0.8191          | 0.6252 |
| 0.9567        | 1.9476 | 800  | 0.7955          | 0.6124 |
| 0.9354        | 2.1900 | 900  | 0.7705          | 0.6046 |
| 0.9037        | 2.4336 | 1000 | 0.7526          | 0.5982 |
| 0.901         | 2.6772 | 1100 | 0.7370          | 0.5960 |
| 0.8888        | 2.9208 | 1200 | 0.7251          | 0.5878 |
| 0.8686        | 3.1632 | 1300 | 0.7125          | 0.5834 |
| 0.8681        | 3.4068 | 1400 | 0.7030          | 0.5770 |
| 0.8428        | 3.6504 | 1500 | 0.6939          | 0.5729 |
| 0.8372        | 3.8940 | 1600 | 0.6849          | 0.5707 |
| 0.8388        | 4.1364 | 1700 | 0.6779          | 0.5667 |
| 0.8222        | 4.3800 | 1800 | 0.6727          | 0.5621 |
| 0.8289        | 4.6236 | 1900 | 0.6665          | 0.5570 |
| 0.8189        | 4.8672 | 2000 | 0.6623          | 0.5564 |
| 0.8073        | 5.1096 | 2100 | 0.6582          | 0.5535 |
| 0.8           | 5.3532 | 2200 | 0.6532          | 0.5505 |
| 0.8051        | 5.5968 | 2300 | 0.6487          | 0.5461 |
| 0.7897        | 5.8404 | 2400 | 0.6454          | 0.5444 |
| 0.7723        | 6.0828 | 2500 | 0.6421          | 0.5446 |
| 0.7805        | 6.3264 | 2600 | 0.6393          | 0.5413 |
| 0.7974        | 6.5700 | 2700 | 0.6365          | 0.5396 |
| 0.7794        | 6.8136 | 2800 | 0.6344          | 0.5392 |
| 0.7676        | 7.0560 | 2900 | 0.6326          | 0.5389 |
| 0.7627        | 7.2996 | 3000 | 0.6305          | 0.5393 |
| 0.7881        | 7.5432 | 3100 | 0.6282          | 0.5379 |
| 0.7689        | 7.7868 | 3200 | 0.6267          | 0.5342 |
| 0.7784        | 8.0292 | 3300 | 0.6253          | 0.5370 |
| 0.7643        | 8.2728 | 3400 | 0.6245          | 0.5345 |
| 0.7817        | 8.5164 | 3500 | 0.6230          | 0.5351 |
| 0.7508        | 8.7600 | 3600 | 0.6218          | 0.5342 |
| 0.7772        | 9.0049 | 3700 | 0.6209          | 0.5334 |
| 0.7624        | 9.2485 | 3800 | 0.6201          | 0.5328 |
| 0.7694        | 9.4921 | 3900 | 0.6196          | 0.5313 |
| 0.7593        | 9.7357 | 4000 | 0.6194          | 0.5308 |
| 0.7585        | 9.9793 | 4100 | 0.6192          | 0.5316 |


### Framework versions

- Transformers 4.52.0.dev0
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.1