Commit
•
30375fb
1
Parent(s):
a770816
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: hubert-large-ll60k-librispeech-clean-100h-demo-dist
|
| 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 |
+
# hubert-large-ll60k-librispeech-clean-100h-demo-dist
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [facebook/hubert-base-ls960](https://huggingface.co/facebook/hubert-base-ls960) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 2.1035
|
| 18 |
+
- Wer: 0.9769
|
| 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: 4
|
| 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: 500
|
| 44 |
+
- num_epochs: 3.0
|
| 45 |
+
- mixed_precision_training: Native AMP
|
| 46 |
+
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 50 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 51 |
+
| 1.4699 | 0.15 | 100 | 2.0098 | 1.0 |
|
| 52 |
+
| 1.1 | 0.3 | 200 | 1.0910 | 0.9769 |
|
| 53 |
+
| 0.6468 | 0.45 | 300 | 1.5089 | 0.9769 |
|
| 54 |
+
| 0.6928 | 0.61 | 400 | 1.6171 | 1.0 |
|
| 55 |
+
| 0.698 | 0.76 | 500 | 1.2406 | 0.9769 |
|
| 56 |
+
| 1.0461 | 0.91 | 600 | 1.3760 | 1.0 |
|
| 57 |
+
| 0.6363 | 1.06 | 700 | 2.1654 | 1.0 |
|
| 58 |
+
| 0.6743 | 1.21 | 800 | 1.7481 | 0.9769 |
|
| 59 |
+
| 0.565 | 1.36 | 900 | 2.1965 | 1.0 |
|
| 60 |
+
| 0.5761 | 1.51 | 1000 | 1.8223 | 0.9769 |
|
| 61 |
+
| 0.6179 | 1.66 | 1100 | 1.9976 | 0.9769 |
|
| 62 |
+
| 0.5052 | 1.82 | 1200 | 1.5585 | 0.9769 |
|
| 63 |
+
| 0.5434 | 1.97 | 1300 | 2.0349 | 0.9769 |
|
| 64 |
+
| 0.4997 | 2.12 | 1400 | 2.4083 | 0.9769 |
|
| 65 |
+
| 0.489 | 2.27 | 1500 | 2.4164 | 0.9769 |
|
| 66 |
+
| 0.5104 | 2.42 | 1600 | 2.4970 | 0.9769 |
|
| 67 |
+
| 0.5324 | 2.57 | 1700 | 2.3352 | 0.9769 |
|
| 68 |
+
| 0.5207 | 2.72 | 1800 | 2.2009 | 0.9769 |
|
| 69 |
+
| 0.5224 | 2.87 | 1900 | 2.1035 | 0.9769 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- Transformers 4.21.0.dev0
|
| 75 |
+
- Pytorch 1.12.0+cu113
|
| 76 |
+
- Datasets 2.3.2
|
| 77 |
+
- Tokenizers 0.12.1
|