|
|
--- |
|
|
library_name: transformers |
|
|
license: apache-2.0 |
|
|
base_model: facebook/wav2vec2-xls-r-300m |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: wav2vec2-E30_freq |
|
|
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-E30_freq |
|
|
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 1.1100 |
|
|
- Cer: 25.3231 |
|
|
|
|
|
## 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: 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: 50 |
|
|
- num_epochs: 3 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Cer | |
|
|
|:-------------:|:------:|:----:|:---------------:|:-------:| |
|
|
| 36.3999 | 0.1289 | 200 | 5.2146 | 100.0 | |
|
|
| 4.9737 | 0.2579 | 400 | 4.6959 | 100.0 | |
|
|
| 4.8 | 0.3868 | 600 | 4.6382 | 100.0 | |
|
|
| 4.7654 | 0.5158 | 800 | 4.6577 | 100.0 | |
|
|
| 4.7312 | 0.6447 | 1000 | 4.6288 | 100.0 | |
|
|
| 4.6911 | 0.7737 | 1200 | 4.5862 | 100.0 | |
|
|
| 4.6383 | 0.9026 | 1400 | 4.4792 | 98.2672 | |
|
|
| 4.5195 | 1.0316 | 1600 | 4.2400 | 97.1511 | |
|
|
| 3.9204 | 1.1605 | 1800 | 3.1009 | 56.4850 | |
|
|
| 2.9139 | 1.2895 | 2000 | 2.5245 | 45.1539 | |
|
|
| 2.5171 | 1.4184 | 2200 | 2.2244 | 43.6090 | |
|
|
| 2.2497 | 1.5474 | 2400 | 1.9096 | 36.6306 | |
|
|
| 2.0305 | 1.6763 | 2600 | 1.7620 | 34.7568 | |
|
|
| 1.8039 | 1.8053 | 2800 | 1.6661 | 34.1988 | |
|
|
| 1.7213 | 1.9342 | 3000 | 1.5361 | 32.5012 | |
|
|
| 1.6193 | 2.0632 | 3200 | 1.4061 | 30.2690 | |
|
|
| 1.4749 | 2.1921 | 3400 | 1.3094 | 29.5054 | |
|
|
| 1.3899 | 2.3211 | 3600 | 1.2938 | 28.8299 | |
|
|
| 1.3117 | 2.4500 | 3800 | 1.2227 | 27.3672 | |
|
|
| 1.3021 | 2.5790 | 4000 | 1.1946 | 26.9854 | |
|
|
| 1.2421 | 2.7079 | 4200 | 1.1484 | 26.1337 | |
|
|
| 1.1926 | 2.8369 | 4400 | 1.1358 | 25.9575 | |
|
|
| 1.1854 | 2.9658 | 4600 | 1.1100 | 25.3231 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.44.2 |
|
|
- Pytorch 2.4.1+cu121 |
|
|
- Datasets 3.0.1 |
|
|
- Tokenizers 0.19.1 |
|
|
|
