Commit
·
9437b30
1
Parent(s):
83c7e5d
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- common_voice
|
| 7 |
+
model-index:
|
| 8 |
+
- name: wav2vec2-large-xlsr-53-demo-colab
|
| 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-large-xlsr-53-demo-colab
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the common_voice dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 6.7860
|
| 20 |
+
- Wer: 1.1067
|
| 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: 0.0003
|
| 40 |
+
- train_batch_size: 16
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- gradient_accumulation_steps: 2
|
| 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: 1000
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
+
|:-------------:|:------:|:----:|:---------------:|:------:|
|
| 55 |
+
| 8.2273 | 44.42 | 400 | 3.3544 | 1.0 |
|
| 56 |
+
| 0.9228 | 88.84 | 800 | 4.7054 | 1.1601 |
|
| 57 |
+
| 0.1423 | 133.32 | 1200 | 5.9489 | 1.1578 |
|
| 58 |
+
| 0.0751 | 177.74 | 1600 | 5.5939 | 1.1717 |
|
| 59 |
+
| 0.0554 | 222.21 | 2000 | 6.1230 | 1.1717 |
|
| 60 |
+
| 0.0356 | 266.63 | 2400 | 6.2845 | 1.1613 |
|
| 61 |
+
| 0.0288 | 311.11 | 2800 | 6.6109 | 1.2100 |
|
| 62 |
+
| 0.0223 | 355.53 | 3200 | 6.5605 | 1.1299 |
|
| 63 |
+
| 0.0197 | 399.95 | 3600 | 7.1242 | 1.1682 |
|
| 64 |
+
| 0.0171 | 444.42 | 4000 | 7.2452 | 1.1578 |
|
| 65 |
+
| 0.0149 | 488.84 | 4400 | 7.4048 | 1.0684 |
|
| 66 |
+
| 0.0118 | 533.32 | 4800 | 6.6227 | 1.1172 |
|
| 67 |
+
| 0.011 | 577.74 | 5200 | 6.7909 | 1.1566 |
|
| 68 |
+
| 0.0095 | 622.21 | 5600 | 6.8088 | 1.1102 |
|
| 69 |
+
| 0.0077 | 666.63 | 6000 | 7.4451 | 1.1311 |
|
| 70 |
+
| 0.0062 | 711.11 | 6400 | 6.8486 | 1.0777 |
|
| 71 |
+
| 0.0051 | 755.53 | 6800 | 6.8812 | 1.1241 |
|
| 72 |
+
| 0.0051 | 799.95 | 7200 | 6.9987 | 1.1450 |
|
| 73 |
+
| 0.0041 | 844.42 | 7600 | 7.3048 | 1.1323 |
|
| 74 |
+
| 0.0044 | 888.84 | 8000 | 6.6644 | 1.1125 |
|
| 75 |
+
| 0.0031 | 933.32 | 8400 | 6.6298 | 1.1148 |
|
| 76 |
+
| 0.0027 | 977.74 | 8800 | 6.7860 | 1.1067 |
|
| 77 |
+
|
| 78 |
+
|
| 79 |
+
### Framework versions
|
| 80 |
+
|
| 81 |
+
- Transformers 4.11.3
|
| 82 |
+
- Pytorch 1.10.0+cu111
|
| 83 |
+
- Datasets 1.14.0
|
| 84 |
+
- Tokenizers 0.10.3
|