Commit
·
21144a0
1
Parent(s):
7eaefdb
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,110 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- common_voice_11_0
|
| 7 |
+
metrics:
|
| 8 |
+
- wer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: wav2vec2-xls-r-1b-ur
|
| 11 |
+
results:
|
| 12 |
+
- task:
|
| 13 |
+
name: Automatic Speech Recognition
|
| 14 |
+
type: automatic-speech-recognition
|
| 15 |
+
dataset:
|
| 16 |
+
name: common_voice_11_0
|
| 17 |
+
type: common_voice_11_0
|
| 18 |
+
config: ur
|
| 19 |
+
split: test
|
| 20 |
+
args: ur
|
| 21 |
+
metrics:
|
| 22 |
+
- name: Wer
|
| 23 |
+
type: wer
|
| 24 |
+
value: 0.48854134406937133
|
| 25 |
+
---
|
| 26 |
+
|
| 27 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 28 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 29 |
+
|
| 30 |
+
# wav2vec2-xls-r-1b-ur
|
| 31 |
+
|
| 32 |
+
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_11_0 dataset.
|
| 33 |
+
It achieves the following results on the evaluation set:
|
| 34 |
+
- Loss: inf
|
| 35 |
+
- Wer: 0.4885
|
| 36 |
+
|
| 37 |
+
## Model description
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Intended uses & limitations
|
| 42 |
+
|
| 43 |
+
More information needed
|
| 44 |
+
|
| 45 |
+
## Training and evaluation data
|
| 46 |
+
|
| 47 |
+
More information needed
|
| 48 |
+
|
| 49 |
+
## Training procedure
|
| 50 |
+
|
| 51 |
+
### Training hyperparameters
|
| 52 |
+
|
| 53 |
+
The following hyperparameters were used during training:
|
| 54 |
+
- learning_rate: 0.0003
|
| 55 |
+
- train_batch_size: 6
|
| 56 |
+
- eval_batch_size: 6
|
| 57 |
+
- seed: 42
|
| 58 |
+
- distributed_type: multi-GPU
|
| 59 |
+
- num_devices: 2
|
| 60 |
+
- total_train_batch_size: 12
|
| 61 |
+
- total_eval_batch_size: 12
|
| 62 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 63 |
+
- lr_scheduler_type: linear
|
| 64 |
+
- lr_scheduler_warmup_steps: 500
|
| 65 |
+
- num_epochs: 15.0
|
| 66 |
+
- mixed_precision_training: Native AMP
|
| 67 |
+
|
| 68 |
+
### Training results
|
| 69 |
+
|
| 70 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 71 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 72 |
+
| 1.7368 | 0.48 | 300 | inf | 0.8191 |
|
| 73 |
+
| 1.8995 | 0.97 | 600 | inf | 0.7919 |
|
| 74 |
+
| 0.9144 | 1.45 | 900 | inf | 0.7805 |
|
| 75 |
+
| 1.166 | 1.94 | 1200 | inf | 0.7087 |
|
| 76 |
+
| 0.7972 | 2.42 | 1500 | inf | 0.6901 |
|
| 77 |
+
| 0.8604 | 2.9 | 1800 | inf | 0.6446 |
|
| 78 |
+
| 0.6569 | 3.39 | 2100 | inf | 0.6560 |
|
| 79 |
+
| 0.7267 | 3.87 | 2400 | inf | 0.6363 |
|
| 80 |
+
| 0.687 | 4.35 | 2700 | inf | 0.6343 |
|
| 81 |
+
| 0.7143 | 4.84 | 3000 | inf | 0.6176 |
|
| 82 |
+
| 0.5283 | 5.32 | 3300 | inf | 0.6084 |
|
| 83 |
+
| 0.6917 | 5.81 | 3600 | inf | 0.5942 |
|
| 84 |
+
| 0.5396 | 6.29 | 3900 | inf | 0.5988 |
|
| 85 |
+
| 0.5523 | 6.77 | 4200 | inf | 0.5600 |
|
| 86 |
+
| 0.3167 | 7.26 | 4500 | inf | 0.5648 |
|
| 87 |
+
| 0.3176 | 7.74 | 4800 | inf | 0.5424 |
|
| 88 |
+
| 0.3987 | 8.23 | 5100 | inf | 0.5440 |
|
| 89 |
+
| 0.3327 | 8.71 | 5400 | inf | 0.5316 |
|
| 90 |
+
| 0.1936 | 9.19 | 5700 | inf | 0.5285 |
|
| 91 |
+
| 0.4701 | 9.68 | 6000 | inf | 0.5207 |
|
| 92 |
+
| 0.3581 | 10.16 | 6300 | inf | 0.5176 |
|
| 93 |
+
| 0.4038 | 10.65 | 6600 | inf | 0.5259 |
|
| 94 |
+
| 0.2699 | 11.13 | 6900 | inf | 0.5226 |
|
| 95 |
+
| 0.2302 | 11.61 | 7200 | inf | 0.5181 |
|
| 96 |
+
| 0.3275 | 12.1 | 7500 | inf | 0.5202 |
|
| 97 |
+
| 0.3024 | 12.58 | 7800 | inf | 0.5307 |
|
| 98 |
+
| 0.2568 | 13.06 | 8100 | inf | 0.5243 |
|
| 99 |
+
| 0.1641 | 13.55 | 8400 | inf | 0.5073 |
|
| 100 |
+
| 0.2637 | 14.03 | 8700 | inf | 0.5015 |
|
| 101 |
+
| 0.1778 | 14.52 | 9000 | inf | 0.4892 |
|
| 102 |
+
| 0.0874 | 15.0 | 9300 | inf | 0.4885 |
|
| 103 |
+
|
| 104 |
+
|
| 105 |
+
### Framework versions
|
| 106 |
+
|
| 107 |
+
- Transformers 4.27.0.dev0
|
| 108 |
+
- Pytorch 1.13.0
|
| 109 |
+
- Datasets 2.7.1
|
| 110 |
+
- Tokenizers 0.13.2
|