Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / yl4579/starganv2-vc issues and pull requests

#100 - Whisper-to-Normal speech conversion

Issue - State: open - Opened by Patelraj8694 7 months ago

#99 - Not Cloning the voice of Newly trained person

Issue - State: open - Opened by Hammad-Bangash 9 months ago - 1 comment

#97 - My recommendation results are not true

Issue - State: open - Opened by Nanshanelectrician 12 months ago - 3 comments

#96 - A problem about cuda error

Issue - State: closed - Opened by XinchenK 12 months ago

#95 - Question regarding parameter update and evaluation

Issue - State: closed - Opened by paul-reu about 1 year ago - 2 comments

#94 - Combining Multiple Styles

Issue - State: closed - Opened by Respaired about 1 year ago - 1 comment

#93 - How to train it for my own custom dataset?

Issue - State: closed - Opened by Hammad-Bangash about 1 year ago - 1 comment

#92 - Regarding Non-speech Vocal data in a dataset

Issue - State: closed - Opened by Respaired about 1 year ago - 1 comment

#89 - 我在使用自己处理数据时，运行代码报错

Issue - State: closed - Opened by love112358 over 1 year ago - 1 comment

#88 - May I ask if it is possible to use dual encoders and dual classifiers for accent conversion

Issue - State: closed - Opened by hhhuazi over 1 year ago - 4 comments
Labels: discussion

#87 - Can i use batchnormalize to replace instancenormalize

Issue - State: closed - Opened by Arthur8312 over 1 year ago - 1 comment

#86 - Eval mode for ASR model and compatibility with PyTorch > 1.7

Pull Request - State: open - Opened by mayank-git-hub over 1 year ago - 1 comment

#84 - How can I adjust the tensor shape for different data?

Issue - State: closed - Opened by Golesheed over 1 year ago - 1 comment

#83 - 从 OneCycleLR 的参数来看，learing_rate 似乎一直都不变？

Issue - State: closed - Opened by yt605155624 over 1 year ago - 1 comment

#82 - Questions about One-to-Many

Issue - State: closed - Opened by roinhutovo over 1 year ago - 1 comment

#81 - TypeError: guvectorize() missing 1 required positional argument: 'signature'

Issue - State: closed - Opened by yurong58 over 1 year ago - 3 comments

#80 - F0 models training and use

Issue - State: closed - Opened by daozhi0921 over 1 year ago - 1 comment

#79 - What was the pytorch version this model used for training?

Issue - State: closed - Opened by yurong58 over 1 year ago - 1 comment

#78 - cannot reproduce converted audio in demo examples

Issue - State: closed - Opened by cwhstcnn almost 2 years ago - 2 comments

#77 - parallelWaveGAN perform badly

Issue - State: closed - Opened by 1nlplearner almost 2 years ago - 1 comment

#76 - many-to-one question

Issue - State: closed - Opened by 1nlplearner almost 2 years ago - 2 comments

#75 - Pre-computed style vectors

Issue - State: closed - Opened by carlosfranzreb almost 2 years ago - 2 comments

#74 - How to disentangle style and speaker information?

Issue - State: open - Opened by mozykhau almost 2 years ago - 8 comments
Labels: discussion

#73 - question about mean / std，and training time

Issue - State: closed - Opened by yt605155624 almost 2 years ago - 2 comments

#72 - Why is ASR model goes to train mode in the training loop

Issue - State: open - Opened by amitaie almost 2 years ago - 11 comments
Labels: bug

#71 - how to get the word_index_dict.txt in Japanese language?

Issue - State: closed - Opened by lunar333 almost 2 years ago - 1 comment

#69 - Support needed for Teacher Forcing HiFiGAN Vocoder Finetuning

Issue - State: closed - Opened by MuruganR96 almost 2 years ago - 1 comment

#68 - Support needed for Multiple Discriminators Implementation

Issue - State: open - Opened by MuruganR96 almost 2 years ago - 2 comments
Labels: help wanted

#67 - RuntimeError cuDNN when training on custom dataset.

Issue - State: closed - Opened by kynthesis almost 2 years ago - 1 comment

#66 - Training problems

Issue - State: closed - Opened by MMMMichaelzhang almost 2 years ago - 2 comments

#65 - How to train new voices ? How to use them?

Issue - State: closed - Opened by TutajITeraz about 2 years ago - 2 comments

#64 - the role of original(vocoder) in the inference.ipynb?

Issue - State: closed - Opened by rmaster520 about 2 years ago - 1 comment

#63 - Seems to be incompatible with the colab environment

Issue - State: closed - Opened by KJZH001 about 2 years ago - 3 comments

#62 - Live inference on windows

Issue - State: closed - Opened by Chopin68 about 2 years ago - 5 comments

#61 - For song vc what should I do

Issue - State: open - Opened by panxin801 about 2 years ago - 9 comments
Labels: discussion

#60 - fp16_run: true causes NaNs

Issue - State: open - Opened by CrackerHax over 2 years ago
Labels: bug

#59 - Is there pretrained HIFI gan vocoder?

Issue - State: closed - Opened by netuserjun over 2 years ago - 10 comments

#58 - Error during training

Issue - State: closed - Opened by CrackerHax over 2 years ago - 3 comments

#57 - is something wrong with my StarGan loss curve?

Issue - State: closed - Opened by superhg over 2 years ago - 2 comments

#56 - Meldataset.py MEL_PARAMS question

Issue - State: closed - Opened by skol101 over 2 years ago - 1 comment

#55 - F0 model training on 160 mels

Issue - State: closed - Opened by ashfaquekhalid over 2 years ago - 5 comments

#54 - what should domain number if I have 200+ speakers?

Issue - State: closed - Opened by superhg over 2 years ago

#53 - about high samplerate training

Issue - State: closed - Opened by MMMMichaelzhang over 2 years ago - 1 comment

#52 - ASR training data must be same with stargan training data?

Issue - State: closed - Opened by joan126 over 2 years ago - 4 comments

#51 - Any to many vc, how to improve the speech intelligence for arbitrary inputs？

Issue - State: open - Opened by Kristopher-Chen over 2 years ago - 16 comments
Labels: discussion

#50 - Questions about Style diversification loss

Issue - State: closed - Opened by Chevalier1024 over 2 years ago - 1 comment

#49 - What is the Overall Compute Footprint of the Pipeline?

Issue - State: closed - Opened by snakers4 over 2 years ago - 2 comments

#48 - How to improve multilingual singing voice conversion?

Issue - State: closed - Opened by MMMMichaelzhang over 2 years ago - 9 comments

#47 - Latent representation

Issue - State: closed - Opened by meyerdav over 2 years ago - 2 comments

#46 - Update style encoder after deleting mapping network

Issue - State: closed - Opened by Charlottecuc over 2 years ago - 2 comments

#45 - how to infer with my own audio files

Issue - State: closed - Opened by mehmetariftasli over 2 years ago - 1 comment

#44 - A problem about Log norm loss

Issue - State: closed - Opened by vanetspark over 2 years ago - 2 comments

#43 - how to fine-tune model

Issue - State: closed - Opened by MMMMichaelzhang over 2 years ago - 5 comments

#42 - a bug about val_list.txt

Issue - State: closed - Opened by vanetspark over 2 years ago - 2 comments

#41 - A possible bug in meldataset.py

Issue - State: closed - Opened by Charlottecuc over 2 years ago - 1 comment

#40 - how to change pitch

Issue - State: closed - Opened by MMMMichaelzhang over 2 years ago - 3 comments

#39 - pitch is not very correct when converting songs

Issue - State: closed - Opened by MMMMichaelzhang over 2 years ago

#38 - online flow inference

Issue - State: closed - Opened by Kristopher-Chen over 2 years ago - 1 comment

#37 - Duration is ~10% shorter that original

Issue - State: closed - Opened by skol101 almost 3 years ago - 1 comment

#36 - Advise on hyper param change

Issue - State: closed - Opened by skol101 almost 3 years ago - 2 comments

#35 - Which GAN vocoders shall work with what hyper params?

Issue - State: closed - Opened by skol101 almost 3 years ago - 4 comments

#34 - Using fine-tuned PWGan with trained model

Issue - State: closed - Opened by skol101 almost 3 years ago - 1 comment

#33 - Speaker selection and training on full VCTK set

Issue - State: closed - Opened by skol101 almost 3 years ago - 2 comments

#32 - Upsampling dataset?

Issue - State: closed - Opened by skol101 almost 3 years ago - 1 comment

#31 - Training questions: samples and resuming

Issue - State: closed - Opened by skol101 almost 3 years ago - 1 comment

#30 - What's the way to finetune pretrained StarGANv2 and ParallelWaveGAN ?

Issue - State: closed - Opened by skol101 almost 3 years ago - 2 comments

#29 - Is it possible to keep the timbre the same and only change the pitch

Issue - State: closed - Opened by 980202006 almost 3 years ago - 1 comment

#28 - Non-parallel dataset

Issue - State: closed - Opened by kikirizki almost 3 years ago - 1 comment

#27 - Jvs corpus

Pull Request - State: closed - Opened by supikiti almost 3 years ago

#26 - Any reason why using consistency regularization?

Issue - State: closed - Opened by MingjieChen almost 3 years ago - 2 comments

#25 - Muti-GPU support

Issue - State: closed - Opened by supernirmesh almost 3 years ago - 1 comment

#23 - Should I re-train ParallelWaveGan

Issue - State: closed - Opened by kikirizki almost 3 years ago - 1 comment

#22 - training issue

Issue - State: closed - Opened by Kristopher-Chen about 3 years ago - 1 comment

#21 - Inference with noisy source

Issue - State: open - Opened by Charlottecuc about 3 years ago - 24 comments
Labels: discussion

#20 - Training f0_mean IndexError: tuple index out of range

Issue - State: closed - Opened by jerrymatjila about 3 years ago - 1 comment

#19 - remove mapping network in training stage

Issue - State: closed - Opened by dragen1860 about 3 years ago - 1 comment

#18 - load from my own training checkpoint failed

Issue - State: closed - Opened by dragen1860 about 3 years ago - 1 comment

#17 - any tips on own dataset? say chinese content to japanese styel?

Issue - State: closed - Opened by dragen1860 about 3 years ago - 1 comment

#16 - live infernce code?

Issue - State: closed - Opened by dragen1860 about 3 years ago - 1 comment

#15 - Attempting live inference.

Issue - State: closed - Opened by Pathos0925 about 3 years ago - 3 comments

#14 - Some doubt about the optimize step of style encoder

Issue - State: closed - Opened by 980202006 about 3 years ago - 3 comments

#13 - How to inference with unseen speakers?

Issue - State: closed - Opened by Charlottecuc about 3 years ago - 1 comment

#12 - Have you compared your model with PPG-based VC models?

Issue - State: closed - Opened by Charlottecuc about 3 years ago - 2 comments

#11 - Some doubt about the loss

Issue - State: closed - Opened by 980202006 about 3 years ago - 2 comments

#10 - Doubts about sampling rate

Issue - State: closed - Opened by 980202006 about 3 years ago - 7 comments

#9 - Can you provide the code for ASR and F0 network training？

Issue - State: closed - Opened by 980202006 about 3 years ago - 39 comments

#8 - Compatibility with custom vocoder checkpoints?

Issue - State: closed - Opened by Kreevoz about 3 years ago - 7 comments

#7 - Are there any details about the neutral to emotional conversion？

Issue - State: closed - Opened by 980202006 about 3 years ago - 2 comments

#6 - Some doubt about any to any voice conversion

Issue - State: open - Opened by 980202006 about 3 years ago - 98 comments
Labels: discussion

#5 - How much loss model will converge？

Issue - State: closed - Opened by 980202006 about 3 years ago - 4 comments

#4 - Is there any guidance on distributed?

Issue - State: closed - Opened by 980202006 about 3 years ago - 4 comments

#3 - Update README.md

Pull Request - State: closed - Opened by ghost over 3 years ago

#2 - Develop

Pull Request - State: closed - Opened by yl4579 over 3 years ago