|
--- |
|
license: apache-2.0 |
|
base_model: facebook/wav2vec2-base |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- timit_asr |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: wav2vec2-base-timit-fine-tuned |
|
results: |
|
- task: |
|
name: Automatic Speech Recognition |
|
type: automatic-speech-recognition |
|
dataset: |
|
name: timit_asr |
|
type: timit_asr |
|
config: clean |
|
split: test |
|
args: clean |
|
metrics: |
|
- name: Wer |
|
type: wer |
|
value: 0.41728125284530637 |
|
--- |
|
|
|
<!-- 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-base-timit-fine-tuned |
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the timit_asr dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4273 |
|
- Wer: 0.4173 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 1000 |
|
- num_epochs: 20.0 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:-------:|:----:|:---------------:|:------:| |
|
| 3.1618 | 0.8621 | 100 | 3.1117 | 1.0 | |
|
| 2.9798 | 1.7241 | 200 | 2.9736 | 1.0 | |
|
| 2.9144 | 2.5862 | 300 | 2.9075 | 1.0 | |
|
| 2.1714 | 3.4483 | 400 | 2.0945 | 1.0325 | |
|
| 1.1579 | 4.3103 | 500 | 1.0451 | 0.8299 | |
|
| 0.6087 | 5.1724 | 600 | 0.6754 | 0.6441 | |
|
| 0.481 | 6.0345 | 700 | 0.5275 | 0.5761 | |
|
| 0.3072 | 6.8966 | 800 | 0.4836 | 0.5264 | |
|
| 0.332 | 7.7586 | 900 | 0.4403 | 0.5234 | |
|
| 0.1876 | 8.6207 | 1000 | 0.4758 | 0.5222 | |
|
| 0.2232 | 9.4828 | 1100 | 0.4508 | 0.4892 | |
|
| 0.1332 | 10.3448 | 1200 | 0.4394 | 0.4740 | |
|
| 0.1085 | 11.2069 | 1300 | 0.4466 | 0.4621 | |
|
| 0.098 | 12.0690 | 1400 | 0.4230 | 0.4493 | |
|
| 0.1219 | 12.9310 | 1500 | 0.4180 | 0.4460 | |
|
| 0.1021 | 13.7931 | 1600 | 0.4179 | 0.4406 | |
|
| 0.0741 | 14.6552 | 1700 | 0.4113 | 0.4309 | |
|
| 0.0896 | 15.5172 | 1800 | 0.4392 | 0.4308 | |
|
| 0.0492 | 16.3793 | 1900 | 0.4202 | 0.4313 | |
|
| 0.0759 | 17.2414 | 2000 | 0.4348 | 0.4207 | |
|
| 0.0406 | 18.1034 | 2100 | 0.4419 | 0.4205 | |
|
| 0.074 | 18.9655 | 2200 | 0.4306 | 0.4200 | |
|
| 0.0378 | 19.8276 | 2300 | 0.4273 | 0.4173 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.42.0.dev0 |
|
- Pytorch 2.3.0.post300 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
