Hugging Face's logo

Spaces:
nambn0321
/
T5_First_US_Accent
Sleeping

App Files Community
nambn0321 commited on
Commit
cfe84f3
·
verified ·
1 Parent(s): c293d71

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +64 -103
app.py CHANGED
@@ -15,109 +15,70 @@ model = model.to(device)
15
  vocoder = vocoder.to(device)
16
 
17
  # Dummy speaker embedding (or load your real one here)
18
- speaker_embedding = torch.tensor([[-8.4146e-02, -9.6771e-03, 2.6844e-02, 2.2544e-02, -1.8455e-03,
19
- -2.5492e-02, -6.2846e-02, -6.6044e-02, 3.1312e-02, 1.9042e-02,
20
- -7.3039e-02, -8.0007e-02, 4.7080e-02, 3.7199e-02, 3.8119e-02,
21
- 5.2214e-02, 1.1470e-02, 3.5968e-02, 8.1480e-03, 1.9952e-02,
22
- 3.5186e-02, 1.3260e-02, -1.4271e-02, -3.8798e-02, -6.8658e-02,
23
- -1.0121e-02, -5.8041e-02, 7.0087e-03, 2.4180e-02, 2.5093e-02,
24
- -2.9513e-03, 4.1835e-02, 3.4997e-02, -4.7138e-02, 5.1652e-02,
25
- -6.1358e-02, 8.9115e-03, 6.5912e-02, -6.1133e-02, -6.3178e-02,
26
- 2.6463e-02, -6.4097e-02, 2.5865e-02, 3.6212e-02, 1.1775e-02,
27
- -1.1237e-01, -1.9422e-02, 1.1771e-02, -5.1472e-02, 6.1495e-02,
28
- 1.8622e-02, 3.6634e-02, 4.0088e-02, 2.4596e-02, -8.8999e-02,
29
- -7.0578e-03, 4.2418e-02, 3.0962e-02, 3.6715e-02, 9.8362e-03,
30
- 2.2486e-02, -8.9123e-03, -1.5486e-02, 9.6940e-03, 2.6662e-02,
31
- 3.0941e-02, 9.7256e-03, -3.7924e-02, -6.1841e-02, -5.8947e-02,
32
- 1.9625e-02, 1.0872e-02, 2.3126e-02, 9.8068e-03, 4.3716e-02,
33
- 2.3031e-02, 1.6826e-02, 1.5088e-02, -7.1926e-02, -9.1217e-02,
34
- -7.4439e-02, -3.6920e-02, -6.2430e-02, -5.1998e-02, -3.4543e-02,
35
- -7.5438e-02, -4.9039e-02, 6.2118e-02, 1.7150e-02, 5.1786e-02,
36
- 2.8916e-02, -7.0776e-02, 4.1283e-03, -7.0833e-02, 7.7827e-02,
37
- -1.3024e-03, 2.8799e-02, 2.4766e-02, -5.6856e-02, -4.9455e-02,
38
- 4.2941e-02, -6.8559e-02, -8.7198e-02, 5.3497e-02, 3.1263e-02,
39
- -4.4196e-02, 4.5291e-02, 3.5016e-02, 2.7069e-02, 1.5925e-02,
40
- -6.9111e-02, 2.8218e-02, 6.5694e-02, -2.9196e-03, 5.4257e-02,
41
- 3.1966e-02, -5.5521e-02, -2.9675e-02, -5.7311e-02, -2.5660e-02,
42
- 2.5544e-02, -6.7957e-02, 3.2566e-02, 5.1615e-02, -6.6451e-02,
43
- 4.0387e-02, -5.5676e-02, 6.8813e-03, 4.2905e-02, 5.4461e-02,
44
- 4.0467e-02, 9.1374e-03, 1.8873e-02, 4.0569e-02, 2.6208e-02,
45
- -5.7306e-02, -4.2936e-02, 3.0970e-02, -5.9009e-02, -2.8001e-03,
46
- 4.3110e-03, 3.9585e-03, -1.6110e-02, 3.8766e-02, -5.6695e-02,
47
- 3.8432e-02, -8.8127e-03, -4.8471e-02, 6.2207e-02, -1.0577e-01,
48
- 5.1350e-02, -3.8907e-02, -5.3652e-02, 3.5648e-02, 1.5506e-02,
49
- 1.6576e-02, 5.0061e-03, -6.9767e-02, -4.8234e-02, 5.2357e-02,
50
- 2.8605e-02, 3.0886e-03, 1.0158e-02, -5.5175e-02, -2.0156e-02,
51
- 8.5321e-03, 1.9991e-02, 3.8980e-02, 7.9981e-03, 4.3244e-02,
52
- 1.0584e-02, -8.0995e-02, 3.8089e-02, -3.5556e-02, -2.6731e-03,
53
- 7.6711e-03, -4.6591e-02, 1.0168e-02, 2.5794e-02, -6.5743e-02,
54
- 4.1506e-02, -5.1859e-02, 2.2370e-02, 2.5183e-02, -5.7133e-02,
55
- -9.8171e-03, 4.1331e-02, 4.1121e-02, -3.7192e-02, -6.7827e-02,
56
- -6.6048e-03, -5.5967e-03, -6.1243e-02, 1.8434e-02, 1.7639e-02,
57
- 3.2567e-02, 4.3591e-02, 1.1444e-04, 2.1000e-02, -3.3589e-02,
58
- 6.3952e-02, 3.2268e-02, -4.5664e-02, 6.0440e-03, 5.8870e-03,
59
- 1.3482e-02, 3.1084e-02, -4.7126e-02, 4.0494e-02, -3.1627e-02,
60
- 4.4450e-03, -8.1786e-02, -3.9848e-02, 5.6240e-02, -4.5132e-02,
61
- 4.4740e-02, 5.3827e-02, -9.0746e-02, 2.2419e-02, 1.4306e-02,
62
- -9.1269e-02, 2.7554e-02, -6.4524e-02, -2.8340e-02, 4.7677e-02,
63
- 4.6020e-02, 6.4275e-02, 8.6387e-02, 6.4071e-02, 2.6549e-02,
64
- -4.1220e-02, 2.2935e-02, 4.3711e-02, -4.0866e-02, -8.3878e-04,
65
- -2.0668e-02, 4.3676e-02, 1.9026e-02, -5.8537e-04, 3.7715e-02,
66
- -6.8229e-02, 2.5884e-02, -8.7095e-02, 9.5215e-03, -3.3552e-02,
67
- 4.5210e-02, 4.0977e-02, 7.1891e-03, 4.1980e-02, 3.2498e-02,
68
- 4.2608e-02, 3.9835e-02, 9.1516e-03, -4.8366e-02, 1.4572e-02,
69
- 2.0647e-02, 3.6124e-02, 4.6580e-02, -7.7210e-02, 5.7704e-02,
70
- 1.1030e-03, -5.8918e-02, 2.9356e-02, 3.4516e-02, 1.2991e-03,
71
- -1.1909e-02, 4.8185e-02, 3.0447e-02, 4.7657e-03, -4.7582e-02,
72
- 3.2026e-02, -1.8162e-02, 1.0278e-02, -3.2285e-02, -6.1007e-02,
73
- 2.3500e-02, 3.6015e-02, 1.5098e-02, -6.6183e-02, -4.5165e-02,
74
- -2.3547e-03, 2.5537e-02, 3.6404e-02, -7.5865e-02, -6.8920e-02,
75
- 4.4063e-02, 2.2352e-02, 5.3221e-02, -1.1117e-01, 1.6892e-03,
76
- 1.5308e-02, 2.4001e-02, -3.9773e-02, 3.4531e-02, -4.2249e-02,
77
- 3.2091e-02, 1.8124e-02, -8.6720e-04, 1.1351e-02, 1.1819e-02,
78
- 6.4638e-02, 2.8144e-02, 2.3477e-02, -6.5364e-02, 9.0942e-03,
79
- 4.2016e-02, 1.3313e-02, 3.1852e-02, 4.7584e-02, 5.6397e-02,
80
- 2.4306e-02, 2.4805e-02, -3.0570e-02, 5.5471e-02, -9.8115e-02,
81
- -7.9501e-02, -2.4848e-02, 2.5320e-02, -6.8306e-02, 5.7311e-02,
82
- -4.5671e-02, 5.5003e-02, -4.7802e-02, 1.1989e-02, 1.5714e-02,
83
- -5.9569e-02, 2.2274e-02, 6.2575e-02, -6.7901e-02, -1.0904e-01,
84
- -5.9256e-02, 3.2378e-02, 2.7536e-03, 4.5289e-02, -7.6812e-02,
85
- 5.6575e-02, 4.3113e-02, -4.5169e-02, -1.9948e-02, -6.8110e-02,
86
- 3.7796e-02, -5.1603e-02, 1.7293e-02, 3.9350e-04, -9.2822e-02,
87
- 2.9600e-02, 1.8698e-02, 2.5137e-02, -4.4556e-02, -5.5333e-02,
88
- -3.0218e-02, -3.8053e-02, -4.6037e-02, 9.3173e-02, 3.0859e-02,
89
- -4.5461e-02, -6.4138e-03, 3.9415e-02, -6.0368e-02, 2.1082e-02,
90
- -1.0852e-01, 3.1828e-02, -6.7432e-03, 2.7266e-02, 4.0655e-02,
91
- -1.2353e-02, -5.5131e-02, 8.4986e-03, 7.7413e-03, 3.8210e-02,
92
- 1.1482e-02, 4.4694e-02, -7.8239e-02, 2.7954e-02, -1.4465e-02,
93
- -2.1566e-03, 4.4880e-02, 1.1331e-02, 4.6676e-02, 4.9100e-02,
94
- -8.2094e-02, -1.0947e-01, 2.4079e-02, 3.3945e-02, 1.6590e-02,
95
- -7.0771e-02, 4.7745e-02, -6.1351e-02, 3.1324e-02, 4.3053e-02,
96
- 3.6573e-02, 1.3540e-02, -4.3418e-02, 2.7820e-02, 1.8301e-02,
97
- 2.2243e-02, 2.3911e-02, -1.0137e-01, -1.4640e-02, 6.4367e-03,
98
- 2.2727e-03, -5.7253e-02, 9.6726e-03, -2.1960e-02, 7.6282e-03,
99
- 6.4420e-03, 5.0940e-03, -6.7982e-02, -3.8508e-02, -6.1518e-02,
100
- -8.3228e-02, 2.2245e-02, 4.1082e-02, -1.2621e-02, 1.4017e-02,
101
- -3.7110e-02, 2.4853e-03, -7.3434e-02, -1.5334e-01, -4.2328e-02,
102
- -3.7094e-02, 5.9061e-03, 2.3480e-02, -4.8760e-02, -5.4049e-02,
103
- 2.4981e-02, 3.9116e-02, 1.9743e-02, 3.5136e-02, -3.8281e-02,
104
- -4.0026e-02, -1.6648e-02, -2.5015e-02, 4.2738e-02, 3.4282e-02,
105
- 2.5185e-03, 3.5040e-02, 3.0350e-02, -6.4696e-02, -5.7804e-02,
106
- 3.7872e-02, 4.9861e-02, 2.4467e-02, 2.2676e-03, 3.6383e-02,
107
- 4.5123e-02, 2.6333e-02, 8.5096e-03, -6.5880e-03, -1.3365e-02,
108
- 3.4063e-02, 6.1048e-03, 8.9429e-03, 1.9870e-02, -7.0909e-03,
109
- 4.1589e-03, 2.4686e-02, -6.6816e-02, -7.1205e-02, 4.5023e-02,
110
- -1.2020e-02, 1.7312e-02, 7.1944e-02, 2.5880e-02, -5.8930e-02,
111
- 2.1297e-02, 3.8620e-02, -8.1744e-02, 1.2381e-02, 2.5355e-02,
112
- -1.8926e-03, 1.7969e-02, 1.8478e-02, 4.2966e-03, -3.4873e-02,
113
- -6.8647e-02, 6.3062e-03, -5.6329e-03, 5.0244e-02, 7.9961e-02,
114
- -4.6829e-02, -5.0599e-02, 2.6159e-02, 3.4554e-02, 4.3835e-02,
115
- 3.8469e-02, -4.2016e-02, 4.2099e-02, -1.9708e-02, -1.1056e-02,
116
- -4.5883e-02, 1.9459e-02, -4.2683e-02, 2.8083e-02, 2.1005e-02,
117
- -1.3023e-03, -4.5802e-02, -3.6830e-02, -7.2250e-02, -5.6689e-02,
118
- 8.9729e-03, 2.5722e-02, -6.7266e-02, 3.6313e-02, -4.3734e-02,
119
- -1.4877e-02, -6.9325e-02, 4.3192e-02, 5.8052e-02, -1.5820e-02,
120
- 4.2151e-02, -6.9208e-02]])
121
 
122
  def tts_generate(text):
123
  print(f"📝 Input text: {text}")
 
15
  vocoder = vocoder.to(device)
16
 
17
  # Dummy speaker embedding (or load your real one here)
18
+ speaker_embedding = torch.tensor([[-0.0743, -0.0103, 0.0260, 0.0237, 0.0045, -0.0173, -0.0721, -0.0579,
19
+ 0.0374, 0.0206, -0.0648, -0.0665, 0.0259, 0.0414, 0.0323, 0.0512,
20
+ -0.0078, 0.0259, 0.0123, 0.0155, 0.0371, 0.0255, -0.0156, -0.0398,
21
+ -0.0612, -0.0098, -0.0582, -0.0046, 0.0377, 0.0320, -0.0028, 0.0450,
22
+ 0.0136, -0.0471, 0.0584, -0.0672, 0.0124, 0.0591, -0.0767, -0.0775,
23
+ 0.0142, -0.0590, 0.0407, 0.0436, 0.0238, -0.1164, -0.0200, 0.0116,
24
+ -0.0551, 0.0721, 0.0228, 0.0490, 0.0465, 0.0149, -0.0871, -0.0100,
25
+ 0.0324, 0.0294, 0.0441, 0.0122, 0.0189, -0.0091, -0.0154, 0.0116,
26
+ 0.0376, 0.0224, 0.0141, -0.0388, -0.0615, -0.0467, 0.0216, 0.0115,
27
+ 0.0205, 0.0136, 0.0419, 0.0258, 0.0181, 0.0173, -0.0678, -0.0821,
28
+ -0.0862, -0.0480, -0.0566, -0.0387, -0.0345, -0.0636, -0.0453, 0.0519,
29
+ 0.0190, 0.0681, 0.0282, -0.0694, -0.0032, -0.0608, 0.0649, -0.0070,
30
+ 0.0200, 0.0304, -0.0486, -0.0640, 0.0396, -0.1017, -0.0794, 0.0478,
31
+ 0.0425, -0.0547, 0.0486, 0.0480, 0.0169, 0.0227, -0.0807, 0.0313,
32
+ 0.0611, -0.0058, 0.0498, 0.0242, -0.0534, -0.0267, -0.0341, -0.0348,
33
+ 0.0220, -0.0662, 0.0370, 0.0365, -0.0660, 0.0279, -0.0644, 0.0143,
34
+ 0.0326, 0.0500, 0.0300, 0.0072, 0.0336, 0.0345, 0.0276, -0.0646,
35
+ -0.0484, -0.0059, -0.0605, 0.0012, 0.0081, 0.0036, -0.0033, 0.0463,
36
+ -0.0506, 0.0270, -0.0066, -0.0609, 0.0493, -0.1155, 0.0447, -0.0371,
37
+ -0.0567, 0.0285, 0.0146, 0.0203, 0.0108, -0.0639, -0.0762, 0.0279,
38
+ 0.0205, 0.0018, 0.0158, -0.0595, -0.0299, 0.0084, 0.0270, 0.0379,
39
+ 0.0132, 0.0510, 0.0261, -0.0636, 0.0276, -0.0498, 0.0167, 0.0027,
40
+ -0.0372, 0.0067, 0.0527, -0.0707, 0.0391, -0.0644, 0.0172, 0.0347,
41
+ -0.0643, -0.0093, 0.0371, 0.0346, -0.0542, -0.0589, -0.0141, 0.0344,
42
+ -0.0659, 0.0478, 0.0131, 0.0165, 0.0172, 0.0042, 0.0322, -0.0516,
43
+ 0.0523, 0.0285, -0.0554, 0.0056, -0.0021, 0.0150, 0.0391, -0.0400,
44
+ 0.0248, -0.0332, 0.0047, -0.0792, -0.0429, 0.0398, -0.0565, 0.0409,
45
+ 0.0457, -0.0870, 0.0314, 0.0226, -0.0816, 0.0377, -0.0779, -0.0134,
46
+ 0.0412, 0.0425, 0.0585, 0.0799, 0.0527, 0.0279, -0.0557, 0.0240,
47
+ 0.0306, -0.0370, 0.0098, -0.0225, 0.0299, 0.0527, -0.0011, 0.0456,
48
+ -0.0768, 0.0237, -0.0966, 0.0106, -0.0521, 0.0512, 0.0424, 0.0236,
49
+ 0.0301, 0.0044, 0.0502, 0.0307, 0.0095, -0.0570, 0.0166, 0.0166,
50
+ 0.0321, 0.0367, -0.0677, 0.0514, 0.0165, -0.0601, 0.0407, 0.0401,
51
+ 0.0020, 0.0015, 0.0574, 0.0310, -0.0053, -0.0610, 0.0391, -0.0212,
52
+ 0.0271, -0.0256, -0.0613, 0.0301, 0.0564, 0.0209, -0.0815, -0.0544,
53
+ -0.0091, 0.0303, 0.0256, -0.0597, -0.0593, 0.0376, 0.0184, 0.0580,
54
+ -0.1039, 0.0021, 0.0159, 0.0319, -0.0386, 0.0322, -0.0432, 0.0292,
55
+ 0.0096, 0.0047, 0.0127, 0.0264, 0.0627, 0.0366, 0.0212, -0.0772,
56
+ 0.0303, 0.0400, 0.0267, 0.0290, 0.0309, 0.0488, 0.0430, 0.0153,
57
+ -0.0187, 0.0440, -0.0995, -0.0837, -0.0254, 0.0274, -0.0638, 0.0500,
58
+ -0.0568, 0.0611, -0.0643, 0.0084, 0.0148, -0.0675, 0.0311, 0.0652,
59
+ -0.0648, -0.0791, -0.0660, 0.0231, 0.0096, 0.0477, -0.0702, 0.0503,
60
+ 0.0446, -0.0523, -0.0305, -0.0593, 0.0238, -0.0557, 0.0130, 0.0067,
61
+ -0.0756, 0.0354, 0.0289, 0.0261, -0.0466, -0.0584, -0.0441, -0.0355,
62
+ -0.0699, 0.1035, 0.0268, -0.0459, -0.0062, 0.0283, -0.0462, 0.0247,
63
+ -0.1061, 0.0222, -0.0052, 0.0058, 0.0479, -0.0126, -0.0533, 0.0160,
64
+ 0.0042, 0.0476, 0.0133, 0.0263, -0.0822, 0.0167, -0.0129, -0.0026,
65
+ 0.0359, 0.0130, 0.0528, 0.0397, -0.0638, -0.1078, 0.0214, 0.0292,
66
+ 0.0351, -0.0545, 0.0406, -0.0787, 0.0306, 0.0389, 0.0332, 0.0178,
67
+ -0.0405, 0.0238, 0.0087, 0.0140, 0.0397, -0.0856, -0.0334, -0.0002,
68
+ -0.0025, -0.0352, 0.0299, -0.0384, 0.0179, 0.0057, 0.0005, -0.0593,
69
+ -0.0505, -0.0592, -0.0831, 0.0174, 0.0417, -0.0128, 0.0286, -0.0422,
70
+ -0.0141, -0.0779, -0.1574, -0.0493, -0.0533, -0.0075, 0.0274, -0.0474,
71
+ -0.0516, 0.0257, 0.0360, 0.0330, 0.0212, -0.0346, -0.0637, -0.0165,
72
+ -0.0254, 0.0295, 0.0180, 0.0093, 0.0260, 0.0096, -0.0626, -0.0537,
73
+ 0.0172, 0.0479, 0.0311, 0.0023, 0.0482, 0.0456, 0.0232, 0.0089,
74
+ -0.0030, -0.0109, 0.0400, 0.0059, 0.0046, 0.0122, 0.0007, -0.0109,
75
+ 0.0188, -0.0746, -0.0615, 0.0463, -0.0136, 0.0101, 0.0435, 0.0257,
76
+ -0.0516, 0.0282, 0.0218, -0.0788, 0.0135, 0.0192, -0.0027, 0.0225,
77
+ 0.0103, 0.0045, -0.0529, -0.0672, 0.0158, -0.0058, 0.0440, 0.0572,
78
+ -0.0373, -0.0386, 0.0256, 0.0211, 0.0453, 0.0515, -0.0624, 0.0371,
79
+ -0.0205, -0.0121, -0.0542, 0.0136, -0.0411, 0.0284, 0.0219, -0.0009,
80
+ -0.0469, -0.0276, -0.0797, -0.0664, 0.0094, 0.0443, -0.0661, 0.0388,
81
+ -0.0244, -0.0143, -0.0674, 0.0379, 0.0583, -0.0234, 0.0413, -0.0651]]).to(device)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
82
 
83
  def tts_generate(text):
84
  print(f"📝 Input text: {text}")