|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
name: Waynehills-STT-doogie-server |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# Waynehills-STT-doogie-server |
|
|
|
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: 1.7114 |
|
- Wer: 1.0056 |
|
|
|
## 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: 16 |
|
- 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: 1000 |
|
- num_epochs: 60 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:| |
|
| 0.0734 | 1.01 | 100 | 1.7114 | 1.0056 | |
|
| 0.074 | 2.02 | 200 | 1.7114 | 1.0056 | |
|
| 0.0707 | 3.03 | 300 | 1.7114 | 1.0056 | |
|
| 0.0727 | 4.04 | 400 | 1.7114 | 1.0056 | |
|
| 0.076 | 5.05 | 500 | 1.7114 | 1.0056 | |
|
| 0.0713 | 6.06 | 600 | 1.7114 | 1.0056 | |
|
| 0.0709 | 7.07 | 700 | 1.7114 | 1.0056 | |
|
| 0.0721 | 8.08 | 800 | 1.7114 | 1.0056 | |
|
| 0.0761 | 9.09 | 900 | 1.7114 | 1.0056 | |
|
| 0.0724 | 10.1 | 1000 | 1.7114 | 1.0056 | |
|
| 0.0749 | 11.11 | 1100 | 1.7114 | 1.0056 | |
|
| 0.0746 | 12.12 | 1200 | 1.7114 | 1.0056 | |
|
| 0.0756 | 13.13 | 1300 | 1.7114 | 1.0056 | |
|
| 0.0712 | 14.14 | 1400 | 1.7114 | 1.0056 | |
|
| 0.0704 | 15.15 | 1500 | 1.7114 | 1.0056 | |
|
| 0.0698 | 16.16 | 1600 | 1.7114 | 1.0056 | |
|
| 0.0715 | 17.17 | 1700 | 1.7114 | 1.0056 | |
|
| 0.0743 | 18.18 | 1800 | 1.7114 | 1.0056 | |
|
| 0.0743 | 19.19 | 1900 | 1.7114 | 1.0056 | |
|
| 0.0745 | 20.2 | 2000 | 1.7114 | 1.0056 | |
|
| 0.0702 | 21.21 | 2100 | 1.7114 | 1.0056 | |
|
| 0.0717 | 22.22 | 2200 | 1.7114 | 1.0056 | |
|
| 0.0673 | 23.23 | 2300 | 1.7114 | 1.0056 | |
|
| 0.0701 | 24.24 | 2400 | 1.7114 | 1.0056 | |
|
| 0.0754 | 25.25 | 2500 | 1.7114 | 1.0056 | |
|
| 0.0677 | 26.26 | 2600 | 1.7114 | 1.0056 | |
|
| 0.0751 | 27.27 | 2700 | 1.7114 | 1.0056 | |
|
| 0.0828 | 28.28 | 2800 | 1.7114 | 1.0056 | |
|
| 0.0714 | 29.29 | 2900 | 1.7114 | 1.0056 | |
|
| 0.0735 | 30.3 | 3000 | 1.7114 | 1.0056 | |
|
| 0.0724 | 31.31 | 3100 | 1.7114 | 1.0056 | |
|
| 0.0777 | 32.32 | 3200 | 1.7114 | 1.0056 | |
|
| 0.0747 | 33.33 | 3300 | 1.7114 | 1.0056 | |
|
| 0.0724 | 34.34 | 3400 | 1.7114 | 1.0056 | |
|
| 0.0717 | 35.35 | 3500 | 1.7114 | 1.0056 | |
|
| 0.0723 | 36.36 | 3600 | 1.7114 | 1.0056 | |
|
| 0.0797 | 37.37 | 3700 | 1.7114 | 1.0056 | |
|
| 0.0693 | 38.38 | 3800 | 1.7114 | 1.0056 | |
|
| 0.0748 | 39.39 | 3900 | 1.7114 | 1.0056 | |
|
| 0.0739 | 40.4 | 4000 | 1.7114 | 1.0056 | |
|
| 0.0701 | 41.41 | 4100 | 1.7114 | 1.0056 | |
|
| 0.079 | 42.42 | 4200 | 1.7114 | 1.0056 | |
|
| 0.0753 | 43.43 | 4300 | 1.7114 | 1.0056 | |
|
| 0.0707 | 44.44 | 4400 | 1.7114 | 1.0056 | |
|
| 0.0724 | 45.45 | 4500 | 1.7114 | 1.0056 | |
|
| 0.0667 | 46.46 | 4600 | 1.7114 | 1.0056 | |
|
| 0.077 | 47.47 | 4700 | 1.7114 | 1.0056 | |
|
| 0.0716 | 48.48 | 4800 | 1.7114 | 1.0056 | |
|
| 0.0731 | 49.49 | 4900 | 1.7114 | 1.0056 | |
|
| 0.0741 | 50.51 | 5000 | 1.7114 | 1.0056 | |
|
| 0.0705 | 51.52 | 5100 | 1.7114 | 1.0056 | |
|
| 0.0736 | 52.53 | 5200 | 1.7114 | 1.0056 | |
|
| 0.0741 | 53.54 | 5300 | 1.7114 | 1.0056 | |
|
| 0.0721 | 54.55 | 5400 | 1.7114 | 1.0056 | |
|
| 0.074 | 55.56 | 5500 | 1.7114 | 1.0056 | |
|
| 0.071 | 56.57 | 5600 | 1.7114 | 1.0056 | |
|
| 0.0723 | 57.58 | 5700 | 1.7114 | 1.0056 | |
|
| 0.0725 | 58.59 | 5800 | 1.7114 | 1.0056 | |
|
| 0.0746 | 59.6 | 5900 | 1.7114 | 1.0056 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.12.5 |
|
- Pytorch 1.10.0+cu113 |
|
- Datasets 1.17.0 |
|
- Tokenizers 0.10.3 |
|
|
