Hugging Face's logo

ahmedjaved812
/
urdu-tts-phonemes-finetuned-extended

Text-to-Audio
Transformers
Safetensors
speecht5
Generated from Trainer
Model card Files
xet
 Community
ahmedjaved812 commited on
Commit
17bce5e
·
verified ·
1 Parent(s): 4935e7f

Model save

Browse files
Files changed (2) hide show
  1. README.md +116 -0
  2. model.safetensors +1 -1
README.md ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: urdu-tts-phonemes-finetuned-extended
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # urdu-tts-phonemes-finetuned-extended
14
+
15
+ This model was trained from scratch on an unknown dataset.
16
+ It achieves the following results on the evaluation set:
17
+ - Loss: 0.8014
18
+
19
+ ## Model description
20
+
21
+ More information needed
22
+
23
+ ## Intended uses & limitations
24
+
25
+ More information needed
26
+
27
+ ## Training and evaluation data
28
+
29
+ More information needed
30
+
31
+ ## Training procedure
32
+
33
+ ### Training hyperparameters
34
+
35
+ The following hyperparameters were used during training:
36
+ - learning_rate: 5e-06
37
+ - train_batch_size: 6
38
+ - eval_batch_size: 2
39
+ - seed: 42
40
+ - distributed_type: multi-GPU
41
+ - num_devices: 2
42
+ - gradient_accumulation_steps: 4
43
+ - total_train_batch_size: 48
44
+ - total_eval_batch_size: 4
45
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 500
48
+ - num_epochs: 50
49
+ - mixed_precision_training: Native AMP
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss |
54
+ |:-------------:|:-------:|:-----:|:---------------:|
55
+ | 4.6038 | 0.9149 | 500 | 0.9382 |
56
+ | 3.9723 | 1.8289 | 1000 | 0.8982 |
57
+ | 3.8796 | 2.7429 | 1500 | 0.8837 |
58
+ | 3.7655 | 3.6569 | 2000 | 0.8691 |
59
+ | 3.7279 | 4.5709 | 2500 | 0.8611 |
60
+ | 3.6874 | 5.4849 | 3000 | 0.8507 |
61
+ | 3.6253 | 6.3989 | 3500 | 0.8450 |
62
+ | 3.5747 | 7.3129 | 4000 | 0.8403 |
63
+ | 3.5732 | 8.2269 | 4500 | 0.8419 |
64
+ | 3.5595 | 9.1409 | 5000 | 0.8367 |
65
+ | 3.5520 | 10.0549 | 5500 | 0.8339 |
66
+ | 3.5297 | 10.9698 | 6000 | 0.8357 |
67
+ | 3.5303 | 11.8838 | 6500 | 0.8333 |
68
+ | 3.5060 | 12.7978 | 7000 | 0.8260 |
69
+ | 3.4967 | 13.7118 | 7500 | 0.8256 |
70
+ | 3.5028 | 14.6258 | 8000 | 0.8239 |
71
+ | 3.4786 | 15.5398 | 8500 | 0.8211 |
72
+ | 3.4918 | 16.4538 | 9000 | 0.8209 |
73
+ | 3.4901 | 17.3678 | 9500 | 0.8193 |
74
+ | 3.4815 | 18.2818 | 10000 | 0.8204 |
75
+ | 3.4688 | 19.1958 | 10500 | 0.8163 |
76
+ | 3.4426 | 20.1098 | 11000 | 0.8160 |
77
+ | 3.4519 | 21.0238 | 11500 | 0.8157 |
78
+ | 3.4514 | 21.9387 | 12000 | 0.8138 |
79
+ | 3.4341 | 22.8527 | 12500 | 0.8148 |
80
+ | 3.4341 | 23.7667 | 13000 | 0.8114 |
81
+ | 3.4261 | 24.6807 | 13500 | 0.8129 |
82
+ | 3.4276 | 25.5947 | 14000 | 0.8111 |
83
+ | 3.4234 | 26.5087 | 14500 | 0.8086 |
84
+ | 3.4379 | 27.4227 | 15000 | 0.8083 |
85
+ | 3.4171 | 28.3367 | 15500 | 0.8055 |
86
+ | 3.4083 | 29.2507 | 16000 | 0.8067 |
87
+ | 3.4227 | 30.1647 | 16500 | 0.8090 |
88
+ | 3.4050 | 31.0787 | 17000 | 0.8063 |
89
+ | 3.4144 | 31.9936 | 17500 | 0.8072 |
90
+ | 3.3968 | 32.9076 | 18000 | 0.8028 |
91
+ | 3.4001 | 33.8216 | 18500 | 0.8056 |
92
+ | 3.3873 | 34.7356 | 19000 | 0.8032 |
93
+ | 3.4032 | 35.6496 | 19500 | 0.8057 |
94
+ | 3.3936 | 36.5636 | 20000 | 0.8040 |
95
+ | 3.3829 | 37.4776 | 20500 | 0.8036 |
96
+ | 3.3927 | 38.3916 | 21000 | 0.8034 |
97
+ | 3.3895 | 39.3056 | 21500 | 0.8037 |
98
+ | 3.3789 | 40.2196 | 22000 | 0.8027 |
99
+ | 3.3938 | 41.1336 | 22500 | 0.8020 |
100
+ | 3.3907 | 42.0476 | 23000 | 0.8018 |
101
+ | 3.3752 | 42.9625 | 23500 | 0.8004 |
102
+ | 3.3759 | 43.8765 | 24000 | 0.8009 |
103
+ | 3.3807 | 44.7905 | 24500 | 0.8015 |
104
+ | 3.3808 | 45.7045 | 25000 | 0.8030 |
105
+ | 3.3720 | 46.6185 | 25500 | 0.8016 |
106
+ | 3.3851 | 47.5325 | 26000 | 0.8018 |
107
+ | 3.3733 | 48.4465 | 26500 | 0.8007 |
108
+ | 3.3703 | 49.3605 | 27000 | 0.8014 |
109
+
110
+
111
+ ### Framework versions
112
+
113
+ - Transformers 5.0.0
114
+ - Pytorch 2.10.0+cu128
115
+ - Datasets 4.8.3
116
+ - Tokenizers 0.22.2
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:770618cc8ed99a6d0d96f735ffd9dd59081eeebcc2b45cfd58e0daf38eb0728b
3
  size 578345352
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2a3a65fd83da9ee5d59879e9279dfbf563a80479e5abe41aac81ba664f8dad1
3
  size 578345352