issues
search
gitmylo
/
bark-voice-cloning-HuBERT-quantizer
The code for the bark-voicecloning model. Training and inference.
MIT License
671
stars
111
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Added ITA support
#57
gc-pwr
closed
2 weeks ago
1
Train myself
#56
blastbeng
opened
3 months ago
4
Turkish language support
#55
ege-adam
closed
8 months ago
4
Support for Swahili Language
#54
reddeath1
opened
9 months ago
0
Torch compile errors on windows 11
#53
rsxdalv
opened
10 months ago
0
How to Train for Non-Verbal Effects Voice?
#52
boringtaskai
closed
10 months ago
2
Stuck on `Installing Demucs`
#51
kubinka0505
opened
10 months ago
0
Added improved functionality and RestAPI
#50
w4hns1nn
opened
11 months ago
0
Request: Write a tutorial for training quantizers
#49
bobcate
opened
11 months ago
0
How to create a quantizer in a dialect that Bark didn't support?
#48
Naozumi520
closed
1 year ago
2
How To Train Chinese Tokenizer
#47
quicksandznzn
opened
1 year ago
0
[Question] This notebook it's for create a speark with trained semantic model?
#46
Maverick1983
opened
1 year ago
0
No module named 'hubert'
#45
Maverick1983
closed
1 year ago
1
Support for Turkish langauge
#44
Someonetoldme584
opened
1 year ago
0
adding batches to training?
#43
Many0therFunctions
opened
1 year ago
0
Omni-Lingual Quantizer?
#42
Many0therFunctions
opened
1 year ago
2
german-HuBERT-quantizer_14_epoch.pth does not have all meta data
#41
FeU-aKlos
closed
1 year ago
2
HTTPError: 404 Client Error: Not Found for url: https://huggingface.co/GitMylo/bark-voice-cloning/resolve/main/japanese-HuBERT-quantizer_24_epoch.pth
#40
gjin10969
opened
1 year ago
1
`KeyError: 'best_loss'` when testing self-trained model
#39
Mareike-RTY
closed
1 year ago
2
RuntimeError: The size of tensor a (28) must match the size of tensor b (33) at non-singleton dimension 2
#38
mikeymcfish
closed
1 year ago
0
issues in notebook due to fairseq version
#37
YongeBai
opened
1 year ago
2
updating notebooks, previously was getting ModuleNotFoundError: No module named 'hubert'
#36
YongeBai
closed
1 year ago
0
import torch library and converting wav tensor to cuda tensor
#35
alwinaind
closed
1 year ago
0
importing torch library to select device
#34
alwinaind
closed
1 year ago
0
adding capability to use gpu if available for faster preparation
#33
alwinaind
closed
1 year ago
1
Switch fairseq dependency to transformers' Hubert
#32
rsxdalv
opened
1 year ago
1
Support Japanese voice cloing
#31
junwchina
closed
1 year ago
7
Speaker switching
#30
Nisha-Nagaraj
opened
1 year ago
2
Bug fix
#28
BrasD99
closed
1 year ago
0
Make the package installable and fix load_from_checkpoint on CPU
#27
rsxdalv
closed
1 year ago
2
TypeError: unsupported operand type(s) for |: 'type' and 'NoneType'
#26
tangzeq
closed
1 year ago
1
fix: [train] [dataset] fixed data out of order
#25
lowkeywx
closed
1 year ago
7
Work with me on an exciting Speech project
#24
sidroopdaska
opened
1 year ago
0
Problems training a Portuguese quantizer model
#23
Subarasheese
closed
1 year ago
12
pin audiolm-pytorch==1.1.4 for easier install
#22
rsxdalv
closed
1 year ago
1
change semantic vocab size to 500 to fit hubert quantizer
#21
jidanhuang
closed
1 year ago
1
Testing
#20
Smartking1
closed
1 year ago
9
Voice to semantic
#19
huydung179
opened
1 year ago
3
Fine-tune for a certain speaker
#18
renxiangnan
closed
1 year ago
9
How to increase quality?
#17
platform-kit
closed
1 year ago
5
Questions about model size with auto_train
#16
xiabo2011
closed
1 year ago
13
German tokenizer available / overtraining
#15
C0untFloyd
closed
1 year ago
4
ModuleNotFoundError: No module named 'hubert'
#14
Barth44
closed
1 year ago
2
Support for Hindi langauge
#13
abhiprojectz
opened
1 year ago
29
Hubert with portuguese
#12
gab-luz
closed
1 year ago
2
can not find tokenizer.pth in GitMylo/bark-voice-cloning
#11
Autonomof
closed
1 year ago
1
how to achieve sample biden's same level?
#10
friendmine
closed
1 year ago
2
ignore __pycache__
#9
rsxdalv
closed
1 year ago
1
Simpler for users, allow use on cpu
#8
NickAnastasoff
closed
1 year ago
3
semantic.npy
#7
taalua
closed
1 year ago
3
Next