issues
search
jik876
/
hifi-gan
HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis
MIT License
1.92k
stars
506
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Section 4.4 End-to-End Speech Synthesis
#169
freedomtowin
opened
1 month ago
0
Library Issue
#168
avdg-dev
opened
2 months ago
0
This would directly work if training on Colab/Kaggle. Fixed some issues with the deprecated libraries.
#167
thetushargoyal
opened
2 months ago
0
Hardcoded num_mels to 80?
#166
bzp83
opened
3 months ago
4
Obvious harmonics appear in the generated wavs
#165
Ziyi6
closed
4 months ago
0
离线部署问题
#164
20246688
opened
5 months ago
0
MelDataset mel VS mel_loss
#163
nikifori
opened
5 months ago
2
mat1 and mat2 shapes cannot be multiplied (80x513 and 1x513)
#162
Huan-phonetic
opened
5 months ago
2
How to fine tune hifi-gan with transformer.
#161
imrankh46
opened
8 months ago
0
d_loss nonconvergence?
#160
a897456
opened
9 months ago
0
Choice and effect of "segment size"
#159
BenjSta
opened
10 months ago
0
MPD and MSD Two discriminators take up a lot of memory
#158
a897456
opened
10 months ago
1
License of pre-trained models
#157
geliAI
opened
11 months ago
0
How to train HiFi-GAN on the VCTK dataset as its wavs is the style of ‘.flac’
#156
DthdZK
opened
11 months ago
1
For Ready-to-use req.txt for training. (Create new txt file and paste this text, Required python 3.8)
#155
iamshreeji-copy2
opened
11 months ago
0
DiscriminatorP(2)/P(3)...P(11)
#154
a897456
opened
11 months ago
0
(conv_pre): Conv1d(80, 512, kernel_size=(7,), stride=(1,), padding=(3,))
#153
a897456
opened
11 months ago
1
kernel_size=3?
#152
a897456
opened
11 months ago
0
Generate mel-spectrograms in numpy format using Tacotron2 with teacher-forcing.
#151
a897456
opened
12 months ago
0
Can I get the mel_spectrogram through the librosa.feature.melspectrogram instead of the Tacotron2
#150
a897456
opened
1 year ago
0
python environment
#149
simonwindtner
opened
1 year ago
5
Mel_loss
#148
wj-gxy
opened
1 year ago
0
Why we need to finetune on Tacotron output?
#147
JiachuanDENG
opened
1 year ago
1
Develop
#146
lordzuko
opened
1 year ago
0
init_weights has no effect after weight_norm
#145
Andras7
opened
1 year ago
0
How to improve HiFi-GAN output mel spectrogram
#144
schnekk
opened
1 year ago
0
learning loss explosion
#143
ikpark09
opened
1 year ago
0
How to convert the genererator files into .pth format and generate the config file that can be used with tts
#142
arnav-newzera
opened
1 year ago
0
temp
#141
kdoh0914
opened
1 year ago
0
what about this config in 16k
#140
weituotian
opened
1 year ago
0
the training Mel must be 80 channels? I use the other shape, it has no error ,but inference
#139
lunar333
opened
1 year ago
2
output of inference.py seems to have high sample rate
#138
Pked01
opened
1 year ago
2
aliasing artifact, how to fix it?
#137
splinter21
closed
10 months ago
1
Pre-trained Discriminator model
#136
compressor1212
opened
1 year ago
0
Teacher-Forcing How To
#135
SuperJonotron
opened
1 year ago
0
Mel spectrogram npy contents
#134
leandro-gracia-gil
opened
1 year ago
3
AssertionError: 4D tensors expect 4 values for padding
#133
dillfrescott
closed
1 year ago
1
How to improve HiFi-GAN in stream TTS applications?
#132
JohnHerry
closed
1 year ago
3
TypeError: guvectorize() missing 1 required positional argument: 'signature'
#131
baipeng0110
opened
1 year ago
2
Tacotron + HIFI GAN Fine tuned: Sounds distorted.
#130
Mixomo
opened
1 year ago
0
no end audio(slice audio) has poor effect
#129
yyjjww
opened
1 year ago
0
Train/test split used for VCTK data
#128
spun-oliver
opened
1 year ago
0
Spectrogram (image)-to-wav
#127
ahmeftah
opened
2 years ago
2
LJSpeech-1.1/wavs/-0113.wav not found
#126
kienld3049
opened
2 years ago
0
Pretrained Hifi GAN vocoder at 16KHz
#125
narendranp
opened
2 years ago
4
how to generated mel-spectrogram?
#124
Deerzh
opened
2 years ago
0
pickle.UnpicklingError: invalid load key, '{'
#123
rafa6g
closed
2 years ago
2
about mel spec extract
#122
zxj329
opened
2 years ago
0
Output Spectrum has no information from 4k to 11k by using pretrained model (generator_v3)
#121
yugeshav
opened
2 years ago
0
Pause between sentence
#120
chikiuso
opened
2 years ago
3
Next