---
library_name: transformers
license: cc-by-nc-4.0
base_model: mms-meta/mms-zeroshot-300m
tags:
- automatic-speech-recognition
- genbed
- mms
- generated_from_trainer
metrics:
- wer
model-index:
- name: mms-zeroshot-300m-genbed-combined-model
  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-zeroshot-300m-genbed-combined-model

This model is a fine-tuned version of [mms-meta/mms-zeroshot-300m](https://huggingface.co/mms-meta/mms-zeroshot-300m) on the GENBED - BEM dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2768
- Wer: 0.4294

## 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: 0.0003
- train_batch_size: 8
- 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: 100
- num_epochs: 30.0
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Wer    |
|:-------------:|:------:|:----:|:---------------:|:------:|
| No log        | 0.2747 | 200  | 2.2031          | 1.0    |
| No log        | 0.5495 | 400  | 0.4235          | 0.6042 |
| 2.7468        | 0.8242 | 600  | 0.3791          | 0.5722 |
| 2.7468        | 1.0989 | 800  | 0.3588          | 0.5632 |
| 0.5446        | 1.3736 | 1000 | 0.3489          | 0.5446 |
| 0.5446        | 1.6484 | 1200 | 0.3406          | 0.5458 |
| 0.5446        | 1.9231 | 1400 | 0.3326          | 0.5183 |
| 0.4932        | 2.1978 | 1600 | 0.3235          | 0.5237 |
| 0.4932        | 2.4725 | 1800 | 0.3202          | 0.5058 |
| 0.4644        | 2.7473 | 2000 | 0.3204          | 0.504  |
| 0.4644        | 3.0220 | 2200 | 0.3144          | 0.4910 |
| 0.4644        | 3.2967 | 2400 | 0.3076          | 0.4957 |
| 0.4505        | 3.5714 | 2600 | 0.3039          | 0.4751 |
| 0.4505        | 3.8462 | 2800 | 0.3027          | 0.4785 |
| 0.433         | 4.1209 | 3000 | 0.3013          | 0.4682 |
| 0.433         | 4.3956 | 3200 | 0.2961          | 0.4765 |
| 0.433         | 4.6703 | 3400 | 0.2914          | 0.468  |
| 0.4203        | 4.9451 | 3600 | 0.2909          | 0.4656 |
| 0.4203        | 5.2198 | 3800 | 0.2946          | 0.4508 |
| 0.4042        | 5.4945 | 4000 | 0.2901          | 0.4434 |
| 0.4042        | 5.7692 | 4200 | 0.2900          | 0.4463 |
| 0.4042        | 6.0440 | 4400 | 0.2840          | 0.4449 |
| 0.3962        | 6.3187 | 4600 | 0.2820          | 0.4448 |
| 0.3962        | 6.5934 | 4800 | 0.2771          | 0.4326 |
| 0.3881        | 6.8681 | 5000 | 0.2768          | 0.4294 |
| 0.3881        | 7.1429 | 5200 | 0.2787          | 0.4385 |
| 0.3881        | 7.4176 | 5400 | 0.2795          | 0.4309 |
| 0.3784        | 7.6923 | 5600 | 0.2803          | 0.4357 |


### Framework versions

- Transformers 4.46.0.dev0
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.0