Commit
•
0d9392f
1
Parent(s):
8ff8189
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,97 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- librispeech_asr
|
| 7 |
+
model-index:
|
| 8 |
+
- name: hubert-base-libri-clean-ft100h
|
| 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 |
+
# hubert-base-libri-clean-ft100h
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [facebook/hubert-base-ls960](https://huggingface.co/facebook/hubert-base-ls960) on the librispeech_asr dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.1324
|
| 20 |
+
- Wer: 0.1597
|
| 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: 5e-05
|
| 40 |
+
- train_batch_size: 8
|
| 41 |
+
- eval_batch_size: 16
|
| 42 |
+
- seed: 42
|
| 43 |
+
- gradient_accumulation_steps: 2
|
| 44 |
+
- total_train_batch_size: 16
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: linear
|
| 47 |
+
- lr_scheduler_warmup_steps: 1000
|
| 48 |
+
- num_epochs: 5
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 55 |
+
| No log | 0.14 | 250 | 4.1508 | 1.0000 |
|
| 56 |
+
| 4.4345 | 0.28 | 500 | 3.8766 | 1.0000 |
|
| 57 |
+
| 4.4345 | 0.42 | 750 | 3.4376 | 1.0000 |
|
| 58 |
+
| 2.8475 | 0.56 | 1000 | 2.7380 | 1.0 |
|
| 59 |
+
| 2.8475 | 0.7 | 1250 | 0.8803 | 0.6766 |
|
| 60 |
+
| 1.1877 | 0.84 | 1500 | 0.5671 | 0.5102 |
|
| 61 |
+
| 1.1877 | 0.98 | 1750 | 0.4537 | 0.4388 |
|
| 62 |
+
| 0.5802 | 1.12 | 2000 | 0.3566 | 0.3740 |
|
| 63 |
+
| 0.5802 | 1.26 | 2250 | 0.2925 | 0.3209 |
|
| 64 |
+
| 0.4301 | 1.4 | 2500 | 0.2613 | 0.2952 |
|
| 65 |
+
| 0.4301 | 1.54 | 2750 | 0.2363 | 0.2715 |
|
| 66 |
+
| 0.3591 | 1.68 | 3000 | 0.2155 | 0.2552 |
|
| 67 |
+
| 0.3591 | 1.82 | 3250 | 0.2062 | 0.2418 |
|
| 68 |
+
| 0.3015 | 1.96 | 3500 | 0.1951 | 0.2308 |
|
| 69 |
+
| 0.3015 | 2.1 | 3750 | 0.1842 | 0.2207 |
|
| 70 |
+
| 0.2698 | 2.24 | 4000 | 0.1900 | 0.2112 |
|
| 71 |
+
| 0.2698 | 2.38 | 4250 | 0.1745 | 0.2048 |
|
| 72 |
+
| 0.2561 | 2.52 | 4500 | 0.1718 | 0.2040 |
|
| 73 |
+
| 0.2561 | 2.66 | 4750 | 0.1625 | 0.1939 |
|
| 74 |
+
| 0.2348 | 2.8 | 5000 | 0.1568 | 0.1867 |
|
| 75 |
+
| 0.2348 | 2.94 | 5250 | 0.1517 | 0.1855 |
|
| 76 |
+
| 0.2278 | 3.08 | 5500 | 0.1501 | 0.1807 |
|
| 77 |
+
| 0.2278 | 3.22 | 5750 | 0.1445 | 0.1772 |
|
| 78 |
+
| 0.2166 | 3.36 | 6000 | 0.1422 | 0.1752 |
|
| 79 |
+
| 0.2166 | 3.5 | 6250 | 0.1418 | 0.1741 |
|
| 80 |
+
| 0.2017 | 3.64 | 6500 | 0.1404 | 0.1695 |
|
| 81 |
+
| 0.2017 | 3.78 | 6750 | 0.1356 | 0.1674 |
|
| 82 |
+
| 0.1922 | 3.92 | 7000 | 0.1350 | 0.1688 |
|
| 83 |
+
| 0.1922 | 4.06 | 7250 | 0.1346 | 0.1638 |
|
| 84 |
+
| 0.1979 | 4.2 | 7500 | 0.1359 | 0.1638 |
|
| 85 |
+
| 0.1979 | 4.34 | 7750 | 0.1336 | 0.1612 |
|
| 86 |
+
| 0.1836 | 4.48 | 8000 | 0.1324 | 0.1613 |
|
| 87 |
+
| 0.1836 | 4.62 | 8250 | 0.1320 | 0.1606 |
|
| 88 |
+
| 0.1891 | 4.76 | 8500 | 0.1325 | 0.1598 |
|
| 89 |
+
| 0.1891 | 4.9 | 8750 | 0.1324 | 0.1597 |
|
| 90 |
+
|
| 91 |
+
|
| 92 |
+
### Framework versions
|
| 93 |
+
|
| 94 |
+
- Transformers 4.17.0
|
| 95 |
+
- Pytorch 1.11.0+cu113
|
| 96 |
+
- Datasets 1.18.3
|
| 97 |
+
- Tokenizers 0.12.1
|