---
base_model: facebook/w2v-bert-2.0
license: mit
datasets:
- thennal/IMaSC
- vrclc/festvox-iiith-ml
- vrclc/openslr63
- smcproject/msc
- mozilla-foundation/common_voice_16_1
metrics:
- wer
tags:
- generated_from_trainer
model-index:
- name: w2v2bert-Malayalam
  results:
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset:
      name: OpenSLR Malayalam -Test
      type: vrclc/openslr63
      config: ml
      split: test
      args: ml
    metrics:
    - type: wer
      value: 20.37
      name: WER
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset:
      name: Goole Fleurs
      type: google/fleurs
      config: ml
      split: test
      args: ml
    metrics:
    - type: wer
      value: 39.27
      name: WER
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset: 
      name: Common Voice 16 Malayalam
      type: mozilla-foundation/common_voice_16_1
      config: ml
      split: test
      args: ml
    metrics:
    - type: wer
      value: 53.14
      name: WER

---


# W2V2-BERT-Malayalam

This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on an these datasets: [IMASC](https://huggingface.co/datasets/thennal/IMaSC), [MSC](https://huggingface.co/datasets/smcproject/MSC), [OpenSLR Malayalam Train split](https://huggingface.co/datasets/vrclc/openslr63), [Festvox Malayalam](https://huggingface.co/datasets/vrclc/festvox-iiith-ml), [common_voice_16_1](https://huggingface.co/datasets/mozilla-foundation/common_voice_16_1) 
It achieves the following results on the evaluation set:
- Loss: 0.1722
- Wer: 0.1299


## Training procedure

Trained on NVIDIA A100 GPU

### 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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Wer    |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 1.1416        | 0.46  | 600   | 0.3393          | 0.4616 |
| 0.1734        | 0.92  | 1200  | 0.2414          | 0.3493 |
| 0.1254        | 1.38  | 1800  | 0.2205          | 0.2963 |
| 0.1097        | 1.84  | 2400  | 0.2157          | 0.3133 |
| 0.0923        | 2.3   | 3000  | 0.1854          | 0.2473 |
| 0.0792        | 2.76  | 3600  | 0.1939          | 0.2471 |
| 0.0696        | 3.22  | 4200  | 0.1720          | 0.2282 |
| 0.0589        | 3.68  | 4800  | 0.1768          | 0.2013 |
| 0.0552        | 4.14  | 5400  | 0.1635          | 0.1864 |
| 0.0437        | 4.6   | 6000  | 0.1501          | 0.1826 |
| 0.0408        | 5.06  | 6600  | 0.1500          | 0.1645 |
| 0.0314        | 5.52  | 7200  | 0.1559          | 0.1655 |
| 0.0317        | 5.98  | 7800  | 0.1448          | 0.1553 |
| 0.022         | 6.44  | 8400  | 0.1592          | 0.1590 |
| 0.0218        | 6.9   | 9000  | 0.1431          | 0.1458 |
| 0.0154        | 7.36  | 9600  | 0.1514          | 0.1366 |
| 0.0141        | 7.82  | 10200 | 0.1540          | 0.1383 |
| 0.0113        | 8.28  | 10800 | 0.1558          | 0.1391 |
| 0.0085        | 8.74  | 11400 | 0.1612          | 0.1356 |
| 0.0072        | 9.2   | 12000 | 0.1697          | 0.1289 |
| 0.0046        | 9.66  | 12600 | 0.1722          | 0.1299 |


### Framework versions

- Transformers 4.39.3
- Pytorch 2.1.1+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1