Commit
•
c18d75a
1
Parent(s):
cb5262c
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- common_voice
|
| 7 |
+
model-index:
|
| 8 |
+
- name: wav2vec2-xls-r-1b-dv
|
| 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-xls-r-1b-dv
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the common_voice dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.1702
|
| 20 |
+
- Wer: 0.2123
|
| 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: 4.5e-05
|
| 40 |
+
- train_batch_size: 8
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- gradient_accumulation_steps: 4
|
| 44 |
+
- total_train_batch_size: 32
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: linear
|
| 47 |
+
- lr_scheduler_warmup_steps: 500
|
| 48 |
+
- num_epochs: 30
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 55 |
+
| 3.8412 | 0.66 | 400 | 0.7160 | 0.7913 |
|
| 56 |
+
| 0.6832 | 1.33 | 800 | 0.3401 | 0.5268 |
|
| 57 |
+
| 0.4624 | 1.99 | 1200 | 0.2671 | 0.4683 |
|
| 58 |
+
| 0.3832 | 2.65 | 1600 | 0.2395 | 0.4410 |
|
| 59 |
+
| 0.3443 | 3.32 | 2000 | 0.2410 | 0.4296 |
|
| 60 |
+
| 0.324 | 3.98 | 2400 | 0.2302 | 0.4143 |
|
| 61 |
+
| 0.2934 | 4.64 | 2800 | 0.2402 | 0.4136 |
|
| 62 |
+
| 0.2773 | 5.31 | 3200 | 0.2134 | 0.4088 |
|
| 63 |
+
| 0.2638 | 5.97 | 3600 | 0.2072 | 0.4037 |
|
| 64 |
+
| 0.2479 | 6.63 | 4000 | 0.2036 | 0.3876 |
|
| 65 |
+
| 0.2424 | 7.3 | 4400 | 0.2037 | 0.3767 |
|
| 66 |
+
| 0.2249 | 7.96 | 4800 | 0.1959 | 0.3802 |
|
| 67 |
+
| 0.2169 | 8.62 | 5200 | 0.1943 | 0.3813 |
|
| 68 |
+
| 0.2109 | 9.29 | 5600 | 0.1944 | 0.3691 |
|
| 69 |
+
| 0.1991 | 9.95 | 6000 | 0.1870 | 0.3589 |
|
| 70 |
+
| 0.1917 | 10.61 | 6400 | 0.1834 | 0.3485 |
|
| 71 |
+
| 0.1862 | 11.28 | 6800 | 0.1857 | 0.3486 |
|
| 72 |
+
| 0.1744 | 11.94 | 7200 | 0.1812 | 0.3330 |
|
| 73 |
+
| 0.171 | 12.6 | 7600 | 0.1797 | 0.3436 |
|
| 74 |
+
| 0.1599 | 13.27 | 8000 | 0.1839 | 0.3319 |
|
| 75 |
+
| 0.1597 | 13.93 | 8400 | 0.1737 | 0.3385 |
|
| 76 |
+
| 0.1494 | 14.59 | 8800 | 0.1807 | 0.3239 |
|
| 77 |
+
| 0.1444 | 15.26 | 9200 | 0.1750 | 0.3155 |
|
| 78 |
+
| 0.1382 | 15.92 | 9600 | 0.1705 | 0.3084 |
|
| 79 |
+
| 0.1299 | 16.58 | 10000 | 0.1777 | 0.2999 |
|
| 80 |
+
| 0.1306 | 17.25 | 10400 | 0.1765 | 0.3056 |
|
| 81 |
+
| 0.1239 | 17.91 | 10800 | 0.1676 | 0.2864 |
|
| 82 |
+
| 0.1149 | 18.57 | 11200 | 0.1774 | 0.2861 |
|
| 83 |
+
| 0.1134 | 19.24 | 11600 | 0.1654 | 0.2699 |
|
| 84 |
+
| 0.1101 | 19.9 | 12000 | 0.1621 | 0.2651 |
|
| 85 |
+
| 0.1038 | 20.56 | 12400 | 0.1686 | 0.2610 |
|
| 86 |
+
| 0.1038 | 21.23 | 12800 | 0.1722 | 0.2559 |
|
| 87 |
+
| 0.0988 | 21.89 | 13200 | 0.1708 | 0.2486 |
|
| 88 |
+
| 0.0949 | 22.55 | 13600 | 0.1696 | 0.2453 |
|
| 89 |
+
| 0.0913 | 23.22 | 14000 | 0.1677 | 0.2424 |
|
| 90 |
+
| 0.0879 | 23.88 | 14400 | 0.1640 | 0.2359 |
|
| 91 |
+
| 0.0888 | 24.54 | 14800 | 0.1697 | 0.2347 |
|
| 92 |
+
| 0.0826 | 25.21 | 15200 | 0.1709 | 0.2314 |
|
| 93 |
+
| 0.0819 | 25.87 | 15600 | 0.1679 | 0.2256 |
|
| 94 |
+
| 0.0793 | 26.53 | 16000 | 0.1701 | 0.2214 |
|
| 95 |
+
| 0.0773 | 27.2 | 16400 | 0.1682 | 0.2176 |
|
| 96 |
+
| 0.0783 | 27.86 | 16800 | 0.1685 | 0.2165 |
|
| 97 |
+
| 0.074 | 28.52 | 17200 | 0.1688 | 0.2155 |
|
| 98 |
+
| 0.0753 | 29.19 | 17600 | 0.1695 | 0.2110 |
|
| 99 |
+
| 0.0699 | 29.85 | 18000 | 0.1702 | 0.2123 |
|
| 100 |
+
|
| 101 |
+
|
| 102 |
+
### Framework versions
|
| 103 |
+
|
| 104 |
+
- Transformers 4.17.0.dev0
|
| 105 |
+
- Pytorch 1.10.2+cu102
|
| 106 |
+
- Datasets 1.18.3
|
| 107 |
+
- Tokenizers 0.11.0
|