Commit
·
0f39d20
1
Parent(s):
abb6d2a
Upload README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: xlsr-wav2vec2-1
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# xlsr-wav2vec2-1
|
| 14 |
+
|
| 15 |
+
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.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 0.5437
|
| 18 |
+
- Wer: 0.4412
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 0.0003
|
| 38 |
+
- train_batch_size: 8
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- gradient_accumulation_steps: 2
|
| 42 |
+
- total_train_batch_size: 16
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- lr_scheduler_warmup_steps: 800
|
| 46 |
+
- num_epochs: 30
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 53 |
+
| 5.517 | 1.38 | 400 | 3.0431 | 1.0 |
|
| 54 |
+
| 1.8387 | 2.76 | 800 | 0.6552 | 0.7263 |
|
| 55 |
+
| 0.5971 | 4.14 | 1200 | 0.5308 | 0.5885 |
|
| 56 |
+
| 0.4153 | 5.52 | 1600 | 0.4667 | 0.5551 |
|
| 57 |
+
| 0.3388 | 6.9 | 2000 | 0.4428 | 0.5260 |
|
| 58 |
+
| 0.2803 | 8.28 | 2400 | 0.4915 | 0.5164 |
|
| 59 |
+
| 0.2613 | 9.65 | 2800 | 0.4904 | 0.4988 |
|
| 60 |
+
| 0.237 | 11.03 | 3200 | 0.4998 | 0.5075 |
|
| 61 |
+
| 0.2175 | 12.41 | 3600 | 0.4905 | 0.4983 |
|
| 62 |
+
| 0.1969 | 13.79 | 4000 | 0.4818 | 0.4877 |
|
| 63 |
+
| 0.1932 | 15.17 | 4400 | 0.5578 | 0.5006 |
|
| 64 |
+
| 0.1782 | 16.55 | 4800 | 0.4981 | 0.4949 |
|
| 65 |
+
| 0.1655 | 17.93 | 5200 | 0.4978 | 0.4940 |
|
| 66 |
+
| 0.1505 | 19.31 | 5600 | 0.5360 | 0.4896 |
|
| 67 |
+
| 0.1362 | 20.69 | 6000 | 0.5441 | 0.4709 |
|
| 68 |
+
| 0.1246 | 22.07 | 6400 | 0.5358 | 0.4650 |
|
| 69 |
+
| 0.1117 | 23.45 | 6800 | 0.5513 | 0.4716 |
|
| 70 |
+
| 0.107 | 24.83 | 7200 | 0.5344 | 0.4578 |
|
| 71 |
+
| 0.0963 | 26.21 | 7600 | 0.5073 | 0.4452 |
|
| 72 |
+
| 0.0846 | 27.59 | 8000 | 0.5335 | 0.4497 |
|
| 73 |
+
| 0.0799 | 28.96 | 8400 | 0.5437 | 0.4412 |
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
### Framework versions
|
| 77 |
+
|
| 78 |
+
- Transformers 4.19.2
|
| 79 |
+
- Pytorch 1.11.0+cu113
|
| 80 |
+
- Datasets 2.2.2
|
| 81 |
+
- Tokenizers 0.12.1
|