Commit ·
463264a
1
Parent(s): 1d3a87d
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: wav2vec2-2
|
| 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 |
+
# wav2vec2-2
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 0.9253
|
| 18 |
+
- Wer: 0.8133
|
| 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: 1e-05
|
| 38 |
+
- train_batch_size: 8
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
+
- lr_scheduler_type: linear
|
| 43 |
+
- lr_scheduler_warmup_steps: 400
|
| 44 |
+
- num_epochs: 10
|
| 45 |
+
|
| 46 |
+
### Training results
|
| 47 |
+
|
| 48 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 49 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 50 |
+
| 8.4469 | 0.34 | 200 | 3.7440 | 1.0 |
|
| 51 |
+
| 3.1152 | 0.69 | 400 | 3.3755 | 1.0 |
|
| 52 |
+
| 2.9228 | 1.03 | 600 | 3.0427 | 1.0 |
|
| 53 |
+
| 2.8661 | 1.38 | 800 | 2.9406 | 1.0 |
|
| 54 |
+
| 2.8402 | 1.72 | 1000 | 2.9034 | 1.0 |
|
| 55 |
+
| 2.8301 | 2.07 | 1200 | 2.8850 | 1.0 |
|
| 56 |
+
| 2.8088 | 2.41 | 1400 | 2.8479 | 1.0 |
|
| 57 |
+
| 2.6892 | 2.75 | 1600 | 2.5800 | 1.0 |
|
| 58 |
+
| 2.3249 | 3.1 | 1800 | 2.1310 | 1.0 |
|
| 59 |
+
| 1.9687 | 3.44 | 2000 | 1.7652 | 0.9982 |
|
| 60 |
+
| 1.7338 | 3.79 | 2200 | 1.5430 | 0.9974 |
|
| 61 |
+
| 1.5698 | 4.13 | 2400 | 1.3927 | 0.9985 |
|
| 62 |
+
| 1.4475 | 4.48 | 2600 | 1.3186 | 0.9911 |
|
| 63 |
+
| 1.3764 | 4.82 | 2800 | 1.2406 | 0.9647 |
|
| 64 |
+
| 1.3022 | 5.16 | 3000 | 1.1954 | 0.9358 |
|
| 65 |
+
| 1.2409 | 5.51 | 3200 | 1.1450 | 0.8990 |
|
| 66 |
+
| 1.1989 | 5.85 | 3400 | 1.1107 | 0.8794 |
|
| 67 |
+
| 1.1478 | 6.2 | 3600 | 1.0839 | 0.8667 |
|
| 68 |
+
| 1.106 | 6.54 | 3800 | 1.0507 | 0.8573 |
|
| 69 |
+
| 1.0792 | 6.88 | 4000 | 1.0179 | 0.8463 |
|
| 70 |
+
| 1.0636 | 7.23 | 4200 | 0.9974 | 0.8355 |
|
| 71 |
+
| 1.0224 | 7.57 | 4400 | 0.9757 | 0.8343 |
|
| 72 |
+
| 1.0166 | 7.92 | 4600 | 0.9641 | 0.8261 |
|
| 73 |
+
| 0.9925 | 8.26 | 4800 | 0.9553 | 0.8183 |
|
| 74 |
+
| 0.9934 | 8.61 | 5000 | 0.9466 | 0.8199 |
|
| 75 |
+
| 0.9741 | 8.95 | 5200 | 0.9353 | 0.8172 |
|
| 76 |
+
| 0.9613 | 9.29 | 5400 | 0.9331 | 0.8133 |
|
| 77 |
+
| 0.9714 | 9.64 | 5600 | 0.9272 | 0.8144 |
|
| 78 |
+
| 0.9593 | 9.98 | 5800 | 0.9253 | 0.8133 |
|
| 79 |
+
|
| 80 |
+
|
| 81 |
+
### Framework versions
|
| 82 |
+
|
| 83 |
+
- Transformers 4.19.2
|
| 84 |
+
- Pytorch 1.11.0+cu113
|
| 85 |
+
- Datasets 2.2.2
|
| 86 |
+
- Tokenizers 0.12.1
|