Commit
·
31fc014
1
Parent(s):
bff2abb
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- wer
|
| 7 |
+
model-index:
|
| 8 |
+
- name: ASR_dear_wav2vec2-thai
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# ASR_dear_wav2vec2-thai
|
| 16 |
+
|
| 17 |
+
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.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.3364
|
| 20 |
+
- Wer: 0.3909
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 0.0001
|
| 40 |
+
- train_batch_size: 32
|
| 41 |
+
- eval_batch_size: 16
|
| 42 |
+
- seed: 42
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- lr_scheduler_warmup_steps: 1000
|
| 46 |
+
- num_epochs: 20
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 52 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 53 |
+
| 7.4289 | 0.75 | 1000 | 3.5725 | 1.0000 |
|
| 54 |
+
| 2.2677 | 1.5 | 2000 | 0.7469 | 0.7886 |
|
| 55 |
+
| 0.9445 | 2.24 | 3000 | 0.5423 | 0.6379 |
|
| 56 |
+
| 0.7801 | 2.99 | 4000 | 0.4628 | 0.5895 |
|
| 57 |
+
| 0.6797 | 3.74 | 5000 | 0.4386 | 0.5518 |
|
| 58 |
+
| 0.6187 | 4.49 | 6000 | 0.4137 | 0.5274 |
|
| 59 |
+
| 0.5702 | 5.24 | 7000 | 0.3906 | 0.4903 |
|
| 60 |
+
| 0.5383 | 5.98 | 8000 | 0.3679 | 0.4824 |
|
| 61 |
+
| 0.5059 | 6.73 | 9000 | 0.3627 | 0.4583 |
|
| 62 |
+
| 0.4829 | 7.48 | 10000 | 0.3523 | 0.4535 |
|
| 63 |
+
| 0.4588 | 8.23 | 11000 | 0.3512 | 0.4560 |
|
| 64 |
+
| 0.4381 | 8.98 | 12000 | 0.3442 | 0.4450 |
|
| 65 |
+
| 0.4127 | 9.72 | 13000 | 0.3446 | 0.4358 |
|
| 66 |
+
| 0.4021 | 10.47 | 14000 | 0.3430 | 0.4239 |
|
| 67 |
+
| 0.3866 | 11.22 | 15000 | 0.3357 | 0.4156 |
|
| 68 |
+
| 0.3729 | 11.97 | 16000 | 0.3436 | 0.4127 |
|
| 69 |
+
| 0.3537 | 12.72 | 17000 | 0.3387 | 0.4117 |
|
| 70 |
+
| 0.3483 | 13.46 | 18000 | 0.3344 | 0.4090 |
|
| 71 |
+
| 0.3384 | 14.21 | 19000 | 0.3365 | 0.4001 |
|
| 72 |
+
| 0.3294 | 14.96 | 20000 | 0.3336 | 0.3991 |
|
| 73 |
+
| 0.3218 | 15.71 | 21000 | 0.3401 | 0.4002 |
|
| 74 |
+
| 0.3113 | 16.45 | 22000 | 0.3432 | 0.3976 |
|
| 75 |
+
| 0.3054 | 17.2 | 23000 | 0.3302 | 0.3959 |
|
| 76 |
+
| 0.2976 | 17.95 | 24000 | 0.3358 | 0.3936 |
|
| 77 |
+
| 0.2955 | 18.7 | 25000 | 0.3340 | 0.3930 |
|
| 78 |
+
| 0.2913 | 19.45 | 26000 | 0.3364 | 0.3909 |
|
| 79 |
+
|
| 80 |
+
|
| 81 |
+
### Framework versions
|
| 82 |
+
|
| 83 |
+
- Transformers 4.26.1
|
| 84 |
+
- Pytorch 1.13.1+cu116
|
| 85 |
+
- Datasets 2.10.1
|
| 86 |
+
- Tokenizers 0.13.2
|