Commit ·
4734f8e
1
Parent(s): 3c7b3d3
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- wer
|
| 7 |
+
model-index:
|
| 8 |
+
- name: wav2vec2-base-timit-ms
|
| 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 |
+
# wav2vec2-base-timit-ms
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.7589
|
| 20 |
+
- Wer: 0.3722
|
| 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: 8
|
| 41 |
+
- eval_batch_size: 8
|
| 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: 80
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 52 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 53 |
+
| 4.0088 | 3.7 | 500 | 2.4873 | 1.0 |
|
| 54 |
+
| 1.0451 | 7.41 | 1000 | 0.9286 | 0.5470 |
|
| 55 |
+
| 0.4081 | 11.11 | 1500 | 0.5935 | 0.4397 |
|
| 56 |
+
| 0.2564 | 14.81 | 2000 | 0.6525 | 0.4292 |
|
| 57 |
+
| 0.183 | 18.52 | 2500 | 0.6578 | 0.4486 |
|
| 58 |
+
| 0.1481 | 22.22 | 3000 | 0.6786 | 0.4231 |
|
| 59 |
+
| 0.1299 | 25.93 | 3500 | 0.6660 | 0.4121 |
|
| 60 |
+
| 0.1044 | 29.63 | 4000 | 0.7713 | 0.4209 |
|
| 61 |
+
| 0.0953 | 33.33 | 4500 | 0.6728 | 0.4038 |
|
| 62 |
+
| 0.0746 | 37.04 | 5000 | 0.7122 | 0.4165 |
|
| 63 |
+
| 0.0627 | 40.74 | 5500 | 0.6950 | 0.4126 |
|
| 64 |
+
| 0.0554 | 44.44 | 6000 | 0.8237 | 0.4082 |
|
| 65 |
+
| 0.0494 | 48.15 | 6500 | 0.7311 | 0.3955 |
|
| 66 |
+
| 0.0426 | 51.85 | 7000 | 0.7717 | 0.3899 |
|
| 67 |
+
| 0.0368 | 55.56 | 7500 | 0.7490 | 0.3933 |
|
| 68 |
+
| 0.0315 | 59.26 | 8000 | 0.7056 | 0.3877 |
|
| 69 |
+
| 0.0274 | 62.96 | 8500 | 0.7897 | 0.3850 |
|
| 70 |
+
| 0.0237 | 66.67 | 9000 | 0.7715 | 0.3850 |
|
| 71 |
+
| 0.0223 | 70.37 | 9500 | 0.7774 | 0.3789 |
|
| 72 |
+
| 0.0177 | 74.07 | 10000 | 0.7598 | 0.3744 |
|
| 73 |
+
| 0.0182 | 77.78 | 10500 | 0.7589 | 0.3722 |
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
### Framework versions
|
| 77 |
+
|
| 78 |
+
- Transformers 4.24.0
|
| 79 |
+
- Pytorch 2.0.0+cu118
|
| 80 |
+
- Datasets 1.18.3
|
| 81 |
+
- Tokenizers 0.13.3
|