Hugging Face's logo

rahafvii
/
test8k

TensorBoard
Safetensors
speecht5
Generated from Trainer
Model card Files
xet
 Metrics Community
rahafvii commited on
Commit
6e4be1f
·
verified ·
1 Parent(s): e380812

End of training

Browse files
Files changed (3) hide show
  1. README.md +38 -58
  2. model.safetensors +1 -1
  3. runs/Nov24_01-28-46_DESKTOP-2JMMFKH/events.out.tfevents.1732400927.DESKTOP-2JMMFKH.5632.1 +2 -2
README.md CHANGED
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the common_voice_13_0 dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.5281
21
 
22
  ## Model description
23
 
@@ -36,72 +36,52 @@ More information needed
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
- - learning_rate: 1e-05
40
- - train_batch_size: 8
41
- - eval_batch_size: 4
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: 500
48
- - training_steps: 5000
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss |
54
  |:-------------:|:-----:|:----:|:---------------:|
55
- | 1.1386 | 0.22 | 100 | 0.9446 |
56
- | 0.9909 | 0.43 | 200 | 0.8305 |
57
- | 0.8844 | 0.65 | 300 | 0.7759 |
58
- | 0.7504 | 0.86 | 400 | 0.6585 |
59
- | 0.701 | 1.08 | 500 | 0.6249 |
60
- | 0.6765 | 1.29 | 600 | 0.6073 |
61
- | 0.6682 | 1.51 | 700 | 0.5973 |
62
- | 0.6673 | 1.72 | 800 | 0.5980 |
63
- | 0.6377 | 1.94 | 900 | 0.5897 |
64
- | 0.6529 | 2.15 | 1000 | 0.5818 |
65
- | 0.6312 | 2.37 | 1100 | 0.5764 |
66
- | 0.6202 | 2.58 | 1200 | 0.5706 |
67
- | 0.6201 | 2.8 | 1300 | 0.5683 |
68
- | 0.6124 | 3.01 | 1400 | 0.5722 |
69
- | 0.6211 | 3.23 | 1500 | 0.5613 |
70
- | 0.6105 | 3.44 | 1600 | 0.5613 |
71
- | 0.6222 | 3.66 | 1700 | 0.5554 |
72
- | 0.6186 | 3.88 | 1800 | 0.5563 |
73
- | 0.5954 | 4.09 | 1900 | 0.5522 |
74
- | 0.5905 | 4.31 | 2000 | 0.5580 |
75
- | 0.5915 | 4.52 | 2100 | 0.5490 |
76
- | 0.5994 | 4.74 | 2200 | 0.5469 |
77
- | 0.6001 | 4.95 | 2300 | 0.5438 |
78
- | 0.5827 | 5.17 | 2400 | 0.5430 |
79
- | 0.5979 | 5.38 | 2500 | 0.5461 |
80
- | 0.5862 | 5.6 | 2600 | 0.5420 |
81
- | 0.5875 | 5.81 | 2700 | 0.5411 |
82
- | 0.5828 | 6.03 | 2800 | 0.5420 |
83
- | 0.5777 | 6.24 | 2900 | 0.5403 |
84
- | 0.5823 | 6.46 | 3000 | 0.5375 |
85
- | 0.584 | 6.67 | 3100 | 0.5352 |
86
- | 0.5693 | 6.89 | 3200 | 0.5336 |
87
- | 0.5601 | 7.1 | 3300 | 0.5322 |
88
- | 0.5689 | 7.32 | 3400 | 0.5361 |
89
- | 0.574 | 7.53 | 3500 | 0.5341 |
90
- | 0.5869 | 7.75 | 3600 | 0.5313 |
91
- | 0.5678 | 7.97 | 3700 | 0.5305 |
92
- | 0.5736 | 8.18 | 3800 | 0.5302 |
93
- | 0.5726 | 8.4 | 3900 | 0.5283 |
94
- | 0.5759 | 8.61 | 4000 | 0.5303 |
95
- | 0.5743 | 8.83 | 4100 | 0.5307 |
96
- | 0.5635 | 9.04 | 4200 | 0.5284 |
97
- | 0.573 | 9.26 | 4300 | 0.5305 |
98
- | 0.5845 | 9.47 | 4400 | 0.5294 |
99
- | 0.5698 | 9.69 | 4500 | 0.5246 |
100
- | 0.5608 | 9.9 | 4600 | 0.5264 |
101
- | 0.5672 | 10.12 | 4700 | 0.5290 |
102
- | 0.568 | 10.33 | 4800 | 0.5297 |
103
- | 0.5594 | 10.55 | 4900 | 0.5288 |
104
- | 0.5655 | 10.76 | 5000 | 0.5281 |
105
 
106
 
107
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the common_voice_13_0 dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.5429
21
 
22
  ## Model description
23
 
 
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
+ - learning_rate: 0.0001
40
+ - train_batch_size: 4
41
+ - eval_batch_size: 2
42
  - seed: 42
43
+ - gradient_accumulation_steps: 8
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
+ - training_steps: 3000
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss |
54
  |:-------------:|:-----:|:----:|:---------------:|
55
+ | 0.7829 | 0.76 | 100 | 0.6936 |
56
+ | 0.6812 | 1.53 | 200 | 0.6392 |
57
+ | 0.6544 | 2.29 | 300 | 0.6275 |
58
+ | 0.6337 | 3.05 | 400 | 0.6147 |
59
+ | 0.6214 | 3.81 | 500 | 0.5963 |
60
+ | 0.602 | 4.58 | 600 | 0.5894 |
61
+ | 0.6078 | 5.34 | 700 | 0.5902 |
62
+ | 0.5892 | 6.1 | 800 | 0.5854 |
63
+ | 0.5842 | 6.86 | 900 | 0.5751 |
64
+ | 0.5836 | 7.63 | 1000 | 0.5691 |
65
+ | 0.5712 | 8.39 | 1100 | 0.5722 |
66
+ | 0.5734 | 9.15 | 1200 | 0.5654 |
67
+ | 0.5669 | 9.91 | 1300 | 0.5539 |
68
+ | 0.5575 | 10.68 | 1400 | 0.5629 |
69
+ | 0.5638 | 11.44 | 1500 | 0.5594 |
70
+ | 0.5522 | 12.2 | 1600 | 0.5550 |
71
+ | 0.5585 | 12.96 | 1700 | 0.5515 |
72
+ | 0.5488 | 13.73 | 1800 | 0.5492 |
73
+ | 0.5536 | 14.49 | 1900 | 0.5579 |
74
+ | 0.5353 | 15.25 | 2000 | 0.5533 |
75
+ | 0.5379 | 16.02 | 2100 | 0.5434 |
76
+ | 0.5369 | 16.78 | 2200 | 0.5495 |
77
+ | 0.5375 | 17.54 | 2300 | 0.5441 |
78
+ | 0.5285 | 18.3 | 2400 | 0.5473 |
79
+ | 0.5262 | 19.07 | 2500 | 0.5369 |
80
+ | 0.5242 | 19.83 | 2600 | 0.5464 |
81
+ | 0.5219 | 20.59 | 2700 | 0.5414 |
82
+ | 0.5132 | 21.35 | 2800 | 0.5426 |
83
+ | 0.517 | 22.12 | 2900 | 0.5442 |
84
+ | 0.5097 | 22.88 | 3000 | 0.5429 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
85
 
86
 
87
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cb8793bdc978debf356c907692a6d44d89d32488c086230b824915ef64a80343
3
  size 577789320
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c77082cc5f2aee5963922c448b36f084c79ced6ed09d778d25ec2bac87ac0cc
3
  size 577789320
runs/Nov24_01-28-46_DESKTOP-2JMMFKH/events.out.tfevents.1732400927.DESKTOP-2JMMFKH.5632.1 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b0c3f5052c3a57d196f09bf53086d0e1e69c62e2ced3f1b02b42a24196c53091
3
- size 38538
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5935be183c93f6bf1777a30e812e26e9af0b8e41752c51b448d4384620204c34
3
+ size 40007