|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: kids_phoneme_sm_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. --> |
|
|
|
# kids_phoneme_sm_model |
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.0236 |
|
- Cer: 0.4987 |
|
|
|
## 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: 0.0004 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 30 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Cer | |
|
|:-------------:|:-----:|:-----:|:---------------:|:------:| |
|
| 2.9442 | 0.74 | 500 | 4.9380 | 1.0 | |
|
| 2.8667 | 1.48 | 1000 | 3.3662 | 1.0 | |
|
| 2.8316 | 2.22 | 1500 | 4.4512 | 1.0 | |
|
| 2.8011 | 2.96 | 2000 | 4.1959 | 1.0 | |
|
| 2.7941 | 3.7 | 2500 | 3.2025 | 1.0 | |
|
| 2.5078 | 4.44 | 3000 | 2.1864 | 1.0 | |
|
| 1.8915 | 5.19 | 3500 | 1.6942 | 0.9979 | |
|
| 1.5858 | 5.93 | 4000 | 1.4032 | 0.9707 | |
|
| 1.3097 | 6.67 | 4500 | 1.1950 | 0.9264 | |
|
| 1.134 | 7.41 | 5000 | 1.0407 | 0.8629 | |
|
| 1.0054 | 8.15 | 5500 | 0.9647 | 0.8089 | |
|
| 0.9141 | 8.89 | 6000 | 0.8932 | 0.7713 | |
|
| 0.7902 | 9.63 | 6500 | 0.8355 | 0.7111 | |
|
| 0.7334 | 10.37 | 7000 | 0.8343 | 0.6986 | |
|
| 0.7315 | 11.11 | 7500 | 0.7893 | 0.6806 | |
|
| 0.6443 | 11.85 | 8000 | 0.7572 | 0.6572 | |
|
| 0.5798 | 12.59 | 8500 | 0.7501 | 0.6522 | |
|
| 0.5845 | 13.33 | 9000 | 0.7337 | 0.6166 | |
|
| 0.5366 | 14.07 | 9500 | 0.8090 | 0.6066 | |
|
| 0.5046 | 14.81 | 10000 | 0.7767 | 0.5924 | |
|
| 0.4569 | 15.56 | 10500 | 0.7593 | 0.6074 | |
|
| 0.425 | 16.3 | 11000 | 0.7844 | 0.5832 | |
|
| 0.4421 | 17.04 | 11500 | 0.7757 | 0.5836 | |
|
| 0.3839 | 17.78 | 12000 | 0.8051 | 0.5782 | |
|
| 0.3483 | 18.52 | 12500 | 0.7850 | 0.5715 | |
|
| 0.3499 | 19.26 | 13000 | 0.8381 | 0.5531 | |
|
| 0.3124 | 20.0 | 13500 | 0.7887 | 0.5527 | |
|
| 0.2715 | 20.74 | 14000 | 0.8220 | 0.5581 | |
|
| 0.2823 | 21.48 | 14500 | 0.8489 | 0.5426 | |
|
| 0.257 | 22.22 | 15000 | 0.8818 | 0.5322 | |
|
| 0.2529 | 22.96 | 15500 | 0.9106 | 0.5259 | |
|
| 0.2219 | 23.7 | 16000 | 0.9197 | 0.5184 | |
|
| 0.2003 | 24.44 | 16500 | 0.9177 | 0.5226 | |
|
| 0.202 | 25.19 | 17000 | 0.9586 | 0.5167 | |
|
| 0.1753 | 25.93 | 17500 | 0.9617 | 0.5159 | |
|
| 0.1781 | 26.67 | 18000 | 0.9664 | 0.5063 | |
|
| 0.1619 | 27.41 | 18500 | 1.0026 | 0.5100 | |
|
| 0.16 | 28.15 | 19000 | 1.0088 | 0.4987 | |
|
| 0.1471 | 28.89 | 19500 | 1.0207 | 0.5033 | |
|
| 0.1459 | 29.63 | 20000 | 1.0236 | 0.4987 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.30.1 |
|
- Pytorch 2.0.0 |
|
- Datasets 2.12.0 |
|
- Tokenizers 0.13.3 |
|
|
