Commit
•
0a09185
1
Parent(s):
6e8e636
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
model-index:
|
| 5 |
+
- name: wav2vec2-large-960h-lv60-self-with-wikipedia-lm-timit
|
| 6 |
+
results: []
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 10 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 11 |
+
|
| 12 |
+
# wav2vec2-large-960h-lv60-self-with-wikipedia-lm-timit
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [gxbag/wav2vec2-large-960h-lv60-self-with-wikipedia-lm](https://huggingface.co/gxbag/wav2vec2-large-960h-lv60-self-with-wikipedia-lm) on the None dataset.
|
| 15 |
+
It achieves the following results on the evaluation set:
|
| 16 |
+
- Loss: 0.0889
|
| 17 |
+
- Wer: 0.4976
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
More information needed
|
| 22 |
+
|
| 23 |
+
## Intended uses & limitations
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Training and evaluation data
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training procedure
|
| 32 |
+
|
| 33 |
+
### Training hyperparameters
|
| 34 |
+
|
| 35 |
+
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 0.0003
|
| 37 |
+
- train_batch_size: 16
|
| 38 |
+
- eval_batch_size: 8
|
| 39 |
+
- seed: 42
|
| 40 |
+
- gradient_accumulation_steps: 2
|
| 41 |
+
- total_train_batch_size: 32
|
| 42 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
+
- lr_scheduler_type: linear
|
| 44 |
+
- lr_scheduler_warmup_steps: 1000
|
| 45 |
+
- num_epochs: 30
|
| 46 |
+
- mixed_precision_training: Native AMP
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 51 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 52 |
+
| 7.7911 | 2.02 | 250 | 3.0896 | 1.0 |
|
| 53 |
+
| 1.3854 | 4.03 | 500 | 0.0704 | 0.5052 |
|
| 54 |
+
| 0.1926 | 6.05 | 750 | 0.0678 | 0.5010 |
|
| 55 |
+
| 0.1472 | 8.06 | 1000 | 0.0794 | 0.5157 |
|
| 56 |
+
| 0.1326 | 10.08 | 1250 | 0.0937 | 0.5031 |
|
| 57 |
+
| 0.104 | 12.1 | 1500 | 0.0859 | 0.5055 |
|
| 58 |
+
| 0.0754 | 14.11 | 1750 | 0.0903 | 0.5031 |
|
| 59 |
+
| 0.0624 | 16.13 | 2000 | 0.0927 | 0.5034 |
|
| 60 |
+
| 0.0594 | 18.14 | 2250 | 0.0929 | 0.5016 |
|
| 61 |
+
| 0.057 | 20.16 | 2500 | 0.0873 | 0.5039 |
|
| 62 |
+
| 0.0476 | 22.18 | 2750 | 0.0974 | 0.5055 |
|
| 63 |
+
| 0.0382 | 24.19 | 3000 | 0.0886 | 0.5003 |
|
| 64 |
+
| 0.0329 | 26.21 | 3250 | 0.0832 | 0.4987 |
|
| 65 |
+
| 0.032 | 28.22 | 3500 | 0.0889 | 0.4976 |
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
### Framework versions
|
| 69 |
+
|
| 70 |
+
- Transformers 4.23.0.dev0
|
| 71 |
+
- Pytorch 1.13.0.dev20220624+cu113
|
| 72 |
+
- Datasets 2.5.2.dev0
|
| 73 |
+
- Tokenizers 0.12.1
|