|
--- |
|
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-twi-colab2 |
|
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-twi-colab2 |
|
|
|
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: 0.4517 |
|
- Wer: 0.4211 |
|
|
|
## 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: 200 |
|
- num_epochs: 4 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:| |
|
| 5.9769 | 0.37 | 200 | 0.7091 | 0.6498 | |
|
| 0.8277 | 0.73 | 400 | 0.4803 | 0.4690 | |
|
| 0.7056 | 1.1 | 600 | 0.4752 | 0.4439 | |
|
| 0.6693 | 1.46 | 800 | 0.4622 | 0.4311 | |
|
| 0.6416 | 1.83 | 1000 | 0.4563 | 0.4324 | |
|
| 0.6242 | 2.19 | 1200 | 0.4573 | 0.4268 | |
|
| 0.6226 | 2.56 | 1400 | 0.4520 | 0.4250 | |
|
| 0.5858 | 2.93 | 1600 | 0.4537 | 0.4222 | |
|
| 0.5751 | 3.29 | 1800 | 0.4540 | 0.4227 | |
|
| 0.5912 | 3.66 | 2000 | 0.4517 | 0.4211 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.0.dev0 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.14.4 |
|
- Tokenizers 0.13.3 |
|
|
