jaywalnut310/vits issues and pull requests

#82 - Can anyone explain me what the boundaries are for?

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

#81 - Questions about 48k audio file train

Issue - State: open - Opened by H4ppyB1rd over 2 years ago - 6 comments

#80 - Dev cython ext

Pull Request - State: closed - Opened by payonear over 2 years ago

#79 - 有辦法做到單行指令轉換嗎?

Issue - State: open - Opened by upright2003 over 2 years ago

#78 - Something wrong hapeened to my interpreter?RuntimeError

Issue - State: closed - Opened by redmist328 over 2 years ago - 3 comments

#77 - RuntimeError: The expanded size of the tensor (8192) must match the existing size (448) at non-singleton dimension 1. Target sizes: [1, 8192]. Tensor sizes: [448]

Issue - State: open - Opened by zgfjzs over 2 years ago

#76 - RuntimeError: view_as_complex is only supported for float and double tensors, but got a tensor of scalar type: Half

Issue - State: closed - Opened by shoang22 over 2 years ago - 3 comments

#75 - RuntimeError During Training

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

#74 - Training error

Issue - State: closed - Opened by Eternity231 over 2 years ago - 6 comments

#73 - AssertionError: 4D tensors expect 4 values for padding

Issue - State: closed - Opened by kagura114 over 2 years ago - 3 comments

#72 - Please help me!

Issue - State: open - Opened by Deerzh over 2 years ago - 3 comments

#71 - How to change the speaker when inferencing?

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

#70 - multi-speaker error

Issue - State: open - Opened by sixyang over 2 years ago - 3 comments

#69 - How to get duration of each phoneme.

Issue - State: open - Opened by tuannvhust over 2 years ago - 1 comment

#68 - Fail to run train.py

Issue - State: closed - Opened by l1006986533 over 2 years ago - 2 comments

#67 - [Bug]`epoch_str` should be increased after loaded checkpoint

Issue - State: open - Opened by cantabile-kwok over 2 years ago - 2 comments

#66 - DEBUG numba

Issue - State: open - Opened by Ayushjaiswal12 over 2 years ago - 4 comments

#65 - RuntimeError: "fill_cuda" not implemented for 'ComplexHalf'

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

#64 - Correlation between characters error rate and noisy result

Issue - State: open - Opened by kikirizki over 2 years ago

#63 - windows monotonic_align problem!

Issue - State: open - Opened by SteveTanggithub over 2 years ago - 5 comments

#62 - Zero division error while training a hindi model

Issue - State: open - Opened by tieincred over 2 years ago - 16 comments

#61 - Key error on tilda

Issue - State: open - Opened by danielkrisp over 2 years ago - 1 comment

#60 - no inf checks were recoreded for this optimizer

Issue - State: open - Opened by a2418701192 over 2 years ago

#59 - Can I train on a Vietnamese custom dataset? If yes, can you specify how to do it?

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

#58 - Multi speaker training error

Issue - State: open - Opened by kumdori88 over 2 years ago - 3 comments

#57 - infinite KL Divergence for low resource language data

Issue - State: open - Opened by mehrzadai over 2 years ago

#56 - Help: monotonic_align !

Issue - State: open - Opened by zhengyongchoa almost 3 years ago - 4 comments

#55 - Could you please explain the KL loss in losses.py

Issue - State: closed - Opened by unrea1-sama almost 3 years ago - 2 comments

#54 - about sampler buckets in dataloader

Issue - State: open - Opened by forwiat almost 3 years ago - 1 comment

#53 - Is it possible use same input generate fixed length of wav

Issue - State: closed - Opened by martin3252 almost 3 years ago

#52 - [Bug] 80000 exceeds legal port number range 0-65535

Issue - State: open - Opened by OlaWod almost 3 years ago

#51 - about multi-speaker data

Issue - State: closed - Opened by LH997 almost 3 years ago - 4 comments

#50 - very hard to construct a suitable environment

Issue - State: open - Opened by coderyiyang almost 3 years ago - 1 comment

#49 - Stochastic duration prediction failed for fastspeech2

Issue - State: open - Opened by LEECHOONGHO almost 3 years ago - 4 comments

#48 - why not use multi-scale D and loss NAN

Issue - State: open - Opened by hdmjdp almost 3 years ago

#47 - Can I use SDP for Fastspeech2?

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

#46 - How is the KL loss computed?

Issue - State: open - Opened by cantabile-kwok almost 3 years ago - 7 comments

#45 - Can't train with fp16 on Nvidia RTX3060

Issue - State: open - Opened by thangnvkcn almost 3 years ago - 2 comments

#44 - Training Time

Issue - State: open - Opened by cantabile-kwok about 3 years ago - 4 comments

#43 - Can MultiHeadAttention replace with nn.MultiheadAttention?

Issue - State: open - Opened by lucasjinreal about 3 years ago

#42 - questions about why not convert English words to phone?

Issue - State: open - Opened by lucasjinreal about 3 years ago - 1 comment

#41 - Export the model to onnx format.

Issue - State: open - Opened by mudong0419 about 3 years ago - 24 comments

#40 - The segment_size in paper and code is different?

Issue - State: open - Opened by xinghua-qu about 3 years ago - 1 comment

#39 - Training on GTX2080

Issue - State: open - Opened by liuhaogeng about 3 years ago - 2 comments

#38 - no performance increase for multi-GPU

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

#37 - A simple multi-process version of preprocess.py

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

#36 - What tools do you use to create/split file lists for new dataset?

Issue - State: open - Opened by skol101 about 3 years ago - 1 comment

#35 - any suggestion for onnx exporting?

Issue - State: open - Opened by wkkuniquegmail over 3 years ago - 8 comments

#34 - work around for pytorch stft backward bug.

Pull Request - State: open - Opened by boltzmann-Li over 3 years ago

#33 - Add dataset support for aishell-3?

Issue - State: open - Opened by luohao123 over 3 years ago

#32 - Provide Full Pretrain models ?

Issue - State: open - Opened by chazo1994 over 3 years ago - 5 comments

#31 - poor performance on short phrases

Issue - State: open - Opened by negidius over 3 years ago - 8 comments

#30 - vits_is_good

Issue - State: closed - Opened by MaxMax2016 over 3 years ago - 1 comment

#29 - Pre-trained model link invalid

Issue - State: open - Opened by Approximetal over 3 years ago - 1 comment

#28 - Training time on VCTK.

Issue - State: open - Opened by mudong0419 over 3 years ago - 5 comments

#27 - Question regarding symbols used

Issue - State: open - Opened by akashicMarga over 3 years ago - 2 comments

#26 - Phonemizer is too slow

Issue - State: closed - Opened by Selimonder over 3 years ago - 4 comments

#25 - CPU infer slow

Issue - State: closed - Opened by OnceJune over 3 years ago - 6 comments

#24 - Training on Tesla K80

Issue - State: open - Opened by StuteePatil over 3 years ago - 3 comments

#23 - How to fix the noise during inference time?

Issue - State: closed - Opened by xinghua-qu over 3 years ago - 5 comments

#22 - Leak memory when runing on CPUs

Issue - State: open - Opened by ductho9799 over 3 years ago

#21 - Questions about the 44KHz audio file train.

Issue - State: open - Opened by rlarhk147 over 3 years ago - 3 comments

#20 - Mispronunciation and what is the purpose of the add_blank config ?

Issue - State: closed - Opened by leminhnguyen over 3 years ago - 5 comments

#19 - Can you provide typical loss figures ? KL diverging

Issue - State: open - Opened by godefv over 3 years ago

#18 - Can use character input ?

Issue - State: open - Opened by bino282 over 3 years ago - 2 comments

#17 - A question about "Glow based TTS"

Issue - State: closed - Opened by WelkinYang over 3 years ago

#16 - Is it possible to manually condition pitch and duration like FastPitch ?

Issue - State: closed - Opened by jmasterx over 3 years ago - 1 comment

#15 - Can't train with fp16 on Nvidia P100

Issue - State: open - Opened by 54696d21 over 3 years ago - 8 comments

#14 - Anybody having luck fine tuning?

Issue - State: closed - Opened by TaoTeCha over 3 years ago - 20 comments

#13 - Questions about Loss_dur

Issue - State: open - Opened by Liujingxiu23 over 3 years ago - 18 comments

#12 - some advice about MAS Algorithm

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

#11 - About ceiling for calculating phoneme duration

Issue - State: open - Opened by reppy4620 over 3 years ago - 3 comments

#10 - How to continue fine tuning of model?

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

#9 - Result getting worse when i use ground truth duration.

Issue - State: open - Opened by AlexanderXuan over 3 years ago - 53 comments

#8 - questions about z_p

Issue - State: closed - Opened by Liujingxiu23 over 3 years ago - 1 comment

#7 - what do missing here?

Issue - State: open - Opened by johnfelipe over 3 years ago - 1 comment

#6 - KL Loss is right?

Issue - State: closed - Opened by BridgetteSong over 3 years ago - 17 comments
Labels: good first issue

GitHub / jaywalnut310/vits issues and pull requests