Commit
·
821422e
1
Parent(s):
e1c59af
update model card README.md
Browse files
README.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
|
|
|
|
|
|
| 5 |
model-index:
|
| 6 |
- name: wav2vec2-large-asr-th
|
| 7 |
results: []
|
|
@@ -12,7 +14,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 12 |
|
| 13 |
# wav2vec2-large-asr-th
|
| 14 |
|
| 15 |
-
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
|
| 17 |
## Model description
|
| 18 |
|
|
@@ -43,6 +49,28 @@ The following hyperparameters were used during training:
|
|
| 43 |
- training_steps: 3200
|
| 44 |
- mixed_precision_training: Native AMP
|
| 45 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
### Framework versions
|
| 47 |
|
| 48 |
- Transformers 4.26.1
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- wer
|
| 7 |
model-index:
|
| 8 |
- name: wav2vec2-large-asr-th
|
| 9 |
results: []
|
|
|
|
| 14 |
|
| 15 |
# wav2vec2-large-asr-th
|
| 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: 0.7903
|
| 20 |
+
- Wer: 0.6032
|
| 21 |
+
- Cer: 0.9340
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 49 |
- training_steps: 3200
|
| 50 |
- mixed_precision_training: Native AMP
|
| 51 |
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
|
| 56 |
+
| 3.6817 | 2.2 | 200 | 3.7530 | 1.0 | 1.0000 |
|
| 57 |
+
| 3.5451 | 4.39 | 400 | 3.5861 | 1.0 | 1.0000 |
|
| 58 |
+
| 3.5003 | 6.59 | 600 | 3.5461 | 1.0 | 0.9985 |
|
| 59 |
+
| 2.0313 | 8.79 | 800 | 1.6731 | 0.9712 | 0.9598 |
|
| 60 |
+
| 1.0353 | 10.98 | 1000 | 0.9283 | 0.7608 | 0.9405 |
|
| 61 |
+
| 0.7667 | 13.19 | 1200 | 0.8119 | 0.7033 | 0.9376 |
|
| 62 |
+
| 0.5697 | 15.38 | 1400 | 0.7711 | 0.6636 | 0.9367 |
|
| 63 |
+
| 0.525 | 17.58 | 1600 | 0.7358 | 0.6556 | 0.9364 |
|
| 64 |
+
| 0.4385 | 19.78 | 1800 | 0.7228 | 0.6260 | 0.9353 |
|
| 65 |
+
| 0.4014 | 21.97 | 2000 | 0.7481 | 0.6237 | 0.9350 |
|
| 66 |
+
| 0.3724 | 24.17 | 2200 | 0.7488 | 0.6222 | 0.9343 |
|
| 67 |
+
| 0.3176 | 26.37 | 2400 | 0.7656 | 0.6071 | 0.9343 |
|
| 68 |
+
| 0.3085 | 28.57 | 2600 | 0.7664 | 0.6006 | 0.9341 |
|
| 69 |
+
| 0.2829 | 30.77 | 2800 | 0.7842 | 0.6079 | 0.9341 |
|
| 70 |
+
| 0.2883 | 32.96 | 3000 | 0.7903 | 0.6032 | 0.9340 |
|
| 71 |
+
| 0.2657 | 35.16 | 3200 | 0.7863 | 0.6015 | 0.9339 |
|
| 72 |
+
|
| 73 |
+
|
| 74 |
### Framework versions
|
| 75 |
|
| 76 |
- Transformers 4.26.1
|