update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,103 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- ru
|
| 4 |
+
license: mit
|
| 5 |
+
base_model: microsoft/speecht5_tts
|
| 6 |
+
tags:
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- mozilla-foundation/common_voice_13_0
|
| 10 |
+
model-index:
|
| 11 |
+
- name: SpeechT5 - Russian translit
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# SpeechT5 - Russian translit
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the Common Voice 13 dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.4853
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 1e-05
|
| 42 |
+
- train_batch_size: 8
|
| 43 |
+
- eval_batch_size: 2
|
| 44 |
+
- seed: 42
|
| 45 |
+
- gradient_accumulation_steps: 8
|
| 46 |
+
- total_train_batch_size: 64
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: linear
|
| 49 |
+
- lr_scheduler_warmup_steps: 400
|
| 50 |
+
- training_steps: 2000
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 56 |
+
| 1.0359 | 0.6 | 50 | 0.8176 |
|
| 57 |
+
| 0.8866 | 1.19 | 100 | 0.6899 |
|
| 58 |
+
| 0.787 | 1.79 | 150 | 0.6478 |
|
| 59 |
+
| 0.7477 | 2.38 | 200 | 0.6233 |
|
| 60 |
+
| 0.6734 | 2.98 | 250 | 0.5630 |
|
| 61 |
+
| 0.6216 | 3.58 | 300 | 0.5429 |
|
| 62 |
+
| 0.593 | 4.17 | 350 | 0.5304 |
|
| 63 |
+
| 0.5817 | 4.77 | 400 | 0.5282 |
|
| 64 |
+
| 0.5734 | 5.37 | 450 | 0.5167 |
|
| 65 |
+
| 0.5688 | 5.96 | 500 | 0.5209 |
|
| 66 |
+
| 0.5662 | 6.56 | 550 | 0.5095 |
|
| 67 |
+
| 0.5609 | 7.15 | 600 | 0.5127 |
|
| 68 |
+
| 0.554 | 7.75 | 650 | 0.5041 |
|
| 69 |
+
| 0.5522 | 8.35 | 700 | 0.5038 |
|
| 70 |
+
| 0.5372 | 8.94 | 750 | 0.4984 |
|
| 71 |
+
| 0.5432 | 9.54 | 800 | 0.4995 |
|
| 72 |
+
| 0.5384 | 10.13 | 850 | 0.4971 |
|
| 73 |
+
| 0.5345 | 10.73 | 900 | 0.4981 |
|
| 74 |
+
| 0.5358 | 11.33 | 950 | 0.4942 |
|
| 75 |
+
| 0.5332 | 11.92 | 1000 | 0.4906 |
|
| 76 |
+
| 0.5334 | 12.52 | 1050 | 0.4897 |
|
| 77 |
+
| 0.5301 | 13.11 | 1100 | 0.4914 |
|
| 78 |
+
| 0.5298 | 13.71 | 1150 | 0.4894 |
|
| 79 |
+
| 0.524 | 14.31 | 1200 | 0.4871 |
|
| 80 |
+
| 0.5221 | 14.9 | 1250 | 0.4884 |
|
| 81 |
+
| 0.525 | 15.5 | 1300 | 0.4883 |
|
| 82 |
+
| 0.5232 | 16.1 | 1350 | 0.4866 |
|
| 83 |
+
| 0.5261 | 16.69 | 1400 | 0.4858 |
|
| 84 |
+
| 0.521 | 17.29 | 1450 | 0.4852 |
|
| 85 |
+
| 0.5225 | 17.88 | 1500 | 0.4849 |
|
| 86 |
+
| 0.5219 | 18.48 | 1550 | 0.4860 |
|
| 87 |
+
| 0.5207 | 19.08 | 1600 | 0.4839 |
|
| 88 |
+
| 0.5192 | 19.67 | 1650 | 0.4851 |
|
| 89 |
+
| 0.516 | 20.27 | 1700 | 0.4860 |
|
| 90 |
+
| 0.5186 | 20.86 | 1750 | 0.4811 |
|
| 91 |
+
| 0.5233 | 21.46 | 1800 | 0.4841 |
|
| 92 |
+
| 0.5145 | 22.06 | 1850 | 0.4819 |
|
| 93 |
+
| 0.5159 | 22.65 | 1900 | 0.4822 |
|
| 94 |
+
| 0.5146 | 23.25 | 1950 | 0.4831 |
|
| 95 |
+
| 0.5175 | 23.85 | 2000 | 0.4853 |
|
| 96 |
+
|
| 97 |
+
|
| 98 |
+
### Framework versions
|
| 99 |
+
|
| 100 |
+
- Transformers 4.31.0
|
| 101 |
+
- Pytorch 2.0.1+cu118
|
| 102 |
+
- Datasets 2.14.4
|
| 103 |
+
- Tokenizers 0.13.3
|