issues
search
mazzzystar
/
randomCNN-voice-transfer
Audio style transfer with shallow random parameters CNN.
https://soundcloud.com/mazzzystar/sets/speech-conversion-sample
405
stars
75
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Question
#27
dillfrescott
opened
8 months ago
0
specify n_fft as keyword argument
#26
akniele
closed
1 year ago
1
Any updates to this?
#25
dillfrescott
closed
2 years ago
2
AttributeError: module 'librosa' has no attribute 'output'
#24
thiago-roque07
opened
3 years ago
1
Update requirements.txt
#23
naveenarun
closed
3 years ago
2
RuntimeError: CUDA error: unspecified launch failure
#22
dgiunchi
opened
4 years ago
1
no voice
#21
JunenuJ
opened
4 years ago
0
Bad very long time to train compare the original project DmitryUlyanov
#20
putuoka
opened
4 years ago
0
Question: difference between this repo and others?
#19
DonaldTsang
opened
4 years ago
1
Input output???
#18
sreenivasmrpivot
opened
4 years ago
1
required os and python versions?
#17
marshonhuckleberry
opened
4 years ago
3
can't optimize a non-leaf Tensor
#16
damiansmazurek
opened
4 years ago
1
Fixed problem with output naming
#15
marianbasti
closed
4 years ago
0
More args for parsing
#14
marianbasti
closed
4 years ago
1
Add argument parser
#13
marianbasti
closed
4 years ago
0
Some unclear details.
#12
ysig
opened
4 years ago
1
Is it possible to retain a model of the voice conversion?
#11
HiroshiKawata
closed
4 years ago
2
bad result i git!!
#10
smalissa
closed
5 years ago
0
Phase restoration algorithm
#9
asmekal
closed
5 years ago
2
RuntimeError: CUDA out of memory
#8
ghost
closed
5 years ago
1
import torch error
#7
oracle9i88
closed
5 years ago
7
some problems
#6
Sun-Ziyi
closed
5 years ago
17
Installation Environment
#5
vangogh0318
closed
5 years ago
2
How to Train this repo, can you please eloborate ?
#4
aishweta
closed
5 years ago
4
when I run train, some problems came out:
#3
Sun-Ziyi
closed
5 years ago
9
Why does it have so much noise and vibration, how can we remove it ?
#2
ishandutta2007
closed
5 years ago
10
Is there a way to use this solution without CUDA?
#1
LeandroFerrazMonteiro
closed
6 years ago
2