Hugging Face's logo

ptah23
/
speecht5_finetuned_voxpopuli_es

Text-to-Speech
Transformers
PyTorch
Safetensors
Spanish
speecht5
text-to-audio
Generated from Trainer
Inference Endpoints
Model card Files Community
2
ptah23 commited on
Commit
b5748f6
•
1 Parent(s): d156f1e

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +102 -0
README.md ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - es
4
+ license: mit
5
+ tags:
6
+ - generated_from_trainer
7
+ datasets:
8
+ - facebook/voxpopuli
9
+ model-index:
10
+ - name: speec T5 ES - Peter Gelderbloem
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
+ # speec T5 ES - Peter Gelderbloem
18
+
19
+ This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the Vox Populi ES dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.4400
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 1e-05
41
+ - train_batch_size: 4
42
+ - eval_batch_size: 4
43
+ - seed: 42
44
+ - gradient_accumulation_steps: 8
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: 500
49
+ - training_steps: 4000
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss |
54
+ |:-------------:|:-----:|:----:|:---------------:|
55
+ | 0.6709 | 0.14 | 100 | 0.6107 |
56
+ | 0.624 | 0.28 | 200 | 0.5773 |
57
+ | 0.5921 | 0.43 | 300 | 0.5397 |
58
+ | 0.5504 | 0.57 | 400 | 0.4941 |
59
+ | 0.5289 | 0.71 | 500 | 0.4807 |
60
+ | 0.5236 | 0.85 | 600 | 0.4732 |
61
+ | 0.5194 | 0.99 | 700 | 0.4672 |
62
+ | 0.5095 | 1.13 | 800 | 0.4650 |
63
+ | 0.508 | 1.28 | 900 | 0.4608 |
64
+ | 0.5029 | 1.42 | 1000 | 0.4580 |
65
+ | 0.5029 | 1.56 | 1100 | 0.4569 |
66
+ | 0.4952 | 1.7 | 1200 | 0.4540 |
67
+ | 0.4939 | 1.84 | 1300 | 0.4540 |
68
+ | 0.4924 | 1.99 | 1400 | 0.4523 |
69
+ | 0.4902 | 2.13 | 1500 | 0.4519 |
70
+ | 0.492 | 2.27 | 1600 | 0.4491 |
71
+ | 0.4896 | 2.41 | 1700 | 0.4497 |
72
+ | 0.4874 | 2.55 | 1800 | 0.4482 |
73
+ | 0.4889 | 2.7 | 1900 | 0.4473 |
74
+ | 0.4888 | 2.84 | 2000 | 0.4471 |
75
+ | 0.4902 | 2.98 | 2100 | 0.4461 |
76
+ | 0.4818 | 3.12 | 2200 | 0.4457 |
77
+ | 0.4854 | 3.26 | 2300 | 0.4451 |
78
+ | 0.4886 | 3.4 | 2400 | 0.4429 |
79
+ | 0.4799 | 3.55 | 2500 | 0.4429 |
80
+ | 0.4806 | 3.69 | 2600 | 0.4429 |
81
+ | 0.4777 | 3.83 | 2700 | 0.4416 |
82
+ | 0.4787 | 3.98 | 2800 | 0.4423 |
83
+ | 0.4823 | 4.12 | 2900 | 0.4419 |
84
+ | 0.4822 | 4.26 | 3000 | 0.4409 |
85
+ | 0.4796 | 4.4 | 3100 | 0.4404 |
86
+ | 0.4805 | 4.54 | 3200 | 0.4418 |
87
+ | 0.4795 | 4.69 | 3300 | 0.4409 |
88
+ | 0.4784 | 4.83 | 3400 | 0.4411 |
89
+ | 0.4812 | 4.97 | 3500 | 0.4406 |
90
+ | 0.478 | 5.11 | 3600 | 0.4399 |
91
+ | 0.4766 | 5.25 | 3700 | 0.4396 |
92
+ | 0.4821 | 5.39 | 3800 | 0.4407 |
93
+ | 0.48 | 5.54 | 3900 | 0.4396 |
94
+ | 0.4795 | 5.68 | 4000 | 0.4400 |
95
+
96
+
97
+ ### Framework versions
98
+
99
+ - Transformers 4.30.2
100
+ - Pytorch 1.12.1+cu116
101
+ - Datasets 2.4.0
102
+ - Tokenizers 0.12.1