Hugging Face's logo

Sandiago21
/
speecht5_finetuned_facebook_voxpopuli_french

Text-to-Speech
Transformers
PyTorch
TensorBoard
Safetensors
speecht5
text-to-audio
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
3
Sandiago21 commited on
Commit
c1e13fa
•
1 Parent(s): a34fd8c

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +40 -29
README.md CHANGED
@@ -1,23 +1,24 @@
1
  ---
 
 
2
  license: mit
3
  tags:
4
  - generated_from_trainer
5
  datasets:
6
- - voxpopuli
7
  model-index:
8
- - name: speecht5_finetuned_facebook_voxpopuli_french
9
  results: []
10
- pipeline_tag: text-to-speech
11
  ---
12
 
13
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
  should probably proofread and complete it, then remove this comment. -->
15
 
16
- # speecht5_finetuned_facebook_voxpopuli_french
17
 
18
- This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the voxpopuli dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.4378
21
 
22
  ## Model description
23
 
@@ -37,40 +38,50 @@ More information needed
37
 
38
  The following hyperparameters were used during training:
39
  - learning_rate: 5e-05
40
- - train_batch_size: 8
41
  - eval_batch_size: 8
42
  - seed: 42
43
- - gradient_accumulation_steps: 4
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: 100
48
- - num_epochs: 20
49
 
50
  ### Training results
51
 
52
  | Training Loss | Epoch | Step | Validation Loss |
53
  |:-------------:|:-----:|:-----:|:---------------:|
54
- | 0.4873 | 1.0 | 1584 | 0.4617 |
55
- | 0.4689 | 2.0 | 3168 | 0.4529 |
56
- | 0.4658 | 3.0 | 4752 | 0.4516 |
57
- | 0.4584 | 4.0 | 6336 | 0.4464 |
58
- | 0.4608 | 5.0 | 7920 | 0.4451 |
59
- | 0.4558 | 6.0 | 9504 | 0.4427 |
60
- | 0.4504 | 7.0 | 11088 | 0.4426 |
61
- | 0.4528 | 8.0 | 12672 | 0.4409 |
62
- | 0.4515 | 9.0 | 14256 | 0.4397 |
63
- | 0.4489 | 10.0 | 15840 | 0.4409 |
64
- | 0.4452 | 11.0 | 17424 | 0.4393 |
65
- | 0.4403 | 12.0 | 19008 | 0.4394 |
66
- | 0.4419 | 13.0 | 20592 | 0.4392 |
67
- | 0.441 | 14.0 | 22176 | 0.4378 |
68
- | 0.4418 | 15.0 | 23760 | 0.4376 |
69
- | 0.4373 | 16.0 | 25344 | 0.4382 |
70
- | 0.4405 | 17.0 | 26928 | 0.4367 |
71
- | 0.4377 | 18.0 | 28512 | 0.4374 |
72
- | 0.4351 | 19.0 | 30096 | 0.4376 |
73
- | 0.4371 | 20.0 | 31680 | 0.4378 |
 
 
 
 
 
 
 
 
 
 
74
 
75
 
76
  ### Framework versions
 
1
  ---
2
+ language:
3
+ - fr
4
  license: mit
5
  tags:
6
  - generated_from_trainer
7
  datasets:
8
+ - facebook/voxpopuli
9
  model-index:
10
+ - name: SpeechT5-french
11
  results: []
 
12
  ---
13
 
14
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
  should probably proofread and complete it, then remove this comment. -->
16
 
17
+ # SpeechT5-french
18
 
19
+ This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the VOXPOPULI dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.4379
22
 
23
  ## Model description
24
 
 
38
 
39
  The following hyperparameters were used during training:
40
  - learning_rate: 5e-05
41
+ - train_batch_size: 16
42
  - eval_batch_size: 8
43
  - seed: 42
44
+ - gradient_accumulation_steps: 2
45
  - total_train_batch_size: 32
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 100
49
+ - num_epochs: 30
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss |
54
  |:-------------:|:-----:|:-----:|:---------------:|
55
+ | 0.4872 | 1.0 | 1584 | 0.4663 |
56
+ | 0.4656 | 2.0 | 3168 | 0.4642 |
57
+ | 0.4686 | 3.0 | 4752 | 0.4533 |
58
+ | 0.4576 | 4.0 | 6336 | 0.4479 |
59
+ | 0.4658 | 5.0 | 7920 | 0.4485 |
60
+ | 0.4536 | 6.0 | 9504 | 0.4443 |
61
+ | 0.4559 | 7.0 | 11088 | 0.4426 |
62
+ | 0.449 | 8.0 | 12672 | 0.4410 |
63
+ | 0.4469 | 9.0 | 14256 | 0.4420 |
64
+ | 0.4565 | 10.0 | 15840 | 0.4402 |
65
+ | 0.4428 | 11.0 | 17424 | 0.4470 |
66
+ | 0.4412 | 12.0 | 19008 | 0.4400 |
67
+ | 0.4437 | 13.0 | 20592 | 0.4396 |
68
+ | 0.4395 | 14.0 | 22176 | 0.4385 |
69
+ | 0.4461 | 15.0 | 23760 | 0.4407 |
70
+ | 0.4401 | 16.0 | 25344 | 0.4387 |
71
+ | 0.4407 | 17.0 | 26928 | 0.4379 |
72
+ | 0.4359 | 18.0 | 28512 | 0.4384 |
73
+ | 0.4338 | 19.0 | 30096 | 0.4387 |
74
+ | 0.4326 | 20.0 | 31680 | 0.4381 |
75
+ | 0.4406 | 21.0 | 33264 | 0.4390 |
76
+ | 0.437 | 22.0 | 34848 | 0.4387 |
77
+ | 0.4357 | 23.0 | 36432 | 0.4389 |
78
+ | 0.4309 | 24.0 | 38016 | 0.4387 |
79
+ | 0.441 | 25.0 | 39600 | 0.4379 |
80
+ | 0.4355 | 26.0 | 41184 | 0.4378 |
81
+ | 0.4312 | 27.0 | 42768 | 0.4380 |
82
+ | 0.4328 | 28.0 | 44352 | 0.4388 |
83
+ | 0.4289 | 29.0 | 45936 | 0.4380 |
84
+ | 0.4291 | 30.0 | 47520 | 0.4379 |
85
 
86
 
87
  ### Framework versions