MoonInTheRiver/DiffSinger issues and pull requests

#110 - Training failed

Issue - State: open - Opened by Arseny5 3 months ago - 1 comment

#109 - 处理数据的时候报这个问题

Issue - State: open - Opened by wangjiancheng-123 5 months ago - 1 comment

#108 - English DiffSinger version

Issue - State: open - Opened by leffff 6 months ago

#107 - 云端训练时报错OSError: Unable to synchronously open file (truncated file: eof = 96, sblock->base_addr = 0, stored_eof = 2048)

Issue - State: open - Opened by Gold-code 8 months ago - 1 comment

#105 - batchsize

Issue - State: open - Opened by he-shuwei 9 months ago

#104 - 请问popCS里的歌曲，都是同一个歌手演唱的吗

Issue - State: open - Opened by Aionrichman 9 months ago

#103 - 什么叫声学模型？什么叫唱法模型？这两个模型是什么关系？

Issue - State: open - Opened by liangcaihua 11 months ago - 1 comment

#102 - Multi-GPU training & batchsize problem

Issue - State: open - Opened by X-Drunker 12 months ago

#101 - 我发的数据集申请邮件没人回复

Issue - State: open - Opened by wlf0322 about 1 year ago - 1 comment

#100 - I encountered an error while running DiffSinger on PopCS

Issue - State: closed - Opened by lidaming1 over 1 year ago - 1 comment

#99 - KeyError: 'hop_size'

Issue - State: open - Opened by ghost over 1 year ago - 3 comments

#98 - 为啥我发的数据集申请邮件没人回复有大佬能给我一个数据集下载链接吗

Issue - State: open - Opened by aemnnn over 1 year ago - 1 comment

#97 - the test infer using opencpop dataset isnot working

Issue - State: open - Opened by lvZic over 1 year ago - 1 comment

#96 - README.md fixup

Pull Request - State: closed - Opened by Atomic-Germ over 1 year ago

#95 - 数据集标注

Issue - State: closed - Opened by hsk-yjk over 1 year ago - 2 comments

#94 - Cannot load testset for infer

Issue - State: closed - Opened by jcn199729 over 1 year ago - 2 comments

#93 - hugging spaces are not working

Issue - State: open - Opened by BaptisteGarcin over 1 year ago - 2 comments

#92 - AttributeError: 'LatestModelCheckpoint' object has no attribute '_save_model'

Issue - State: open - Opened by JY9087 over 1 year ago - 1 comment

#91 - Any place to send donations? (aslo a message for gratitude)

Issue - State: closed - Opened by marty1885 over 1 year ago - 2 comments

#90 - Binarize.py error, hparams not accessible from multiprocessing?

Issue - State: closed - Opened by hilmiyafia over 1 year ago - 4 comments

#89 - About Hyperparmeter "predictor_grad"

Issue - State: open - Opened by HAN-oQo over 1 year ago - 2 comments

#88 - Where is your metadata_phone.csv from while running tts?

Issue - State: closed - Opened by saber258 over 1 year ago - 3 comments

#87 - Demo page is not accessible

Issue - State: open - Opened by yewon990 over 1 year ago - 1 comment

#86 - How to calculate spec_min and spec_max?

Issue - State: closed - Opened by ChuhanWang10 over 1 year ago - 3 comments

#85 - Reduce the number of dataset songs

Issue - State: open - Opened by hsk-yjk almost 2 years ago

#84 - 下载ZIP之后显示无法打开，请问有知道的大佬吗？

Issue - State: open - Opened by liulian233-23 almost 2 years ago - 2 comments

#82 - 用A卡怎么训练

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

#81 - IndexError: index 0 is out of bounds for axis 0 with size 0

Issue - State: open - Opened by huajian197 almost 2 years ago - 1 comment

#80 - Please Add requirements_4090.txt File

Issue - State: open - Opened by djt-dev almost 2 years ago - 1 comment

#79 - Inference audio generated at higher speed than training files

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

#78 - AttributeError: 'LatestModelCheckpoint' object has no attribute '_save_model'

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

#77 - RuntimeError: index 155 is out of bounds for dimension 1 with size 155

Issue - State: closed - Opened by ElizavetaSedova almost 2 years ago - 6 comments
Labels: Multi-GPU

#76 - Using the Universal Vocoder

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

#75 - Update README.md

Pull Request - State: open - Opened by eltociear almost 2 years ago

#74 - Hello, I have issue as I try to use another english dataset. And I'm wondering why Inference from packed test set can work (`CUDA_VISIBLE_DEVICES=0 python tasks/run.py --config usr/configs/midi/e2e/opencpop/ds100_adj_rel.yaml --exp_name $MY_DS_EXP_NAME --reset --infer`) but inference model from raw input (`python inference/svs/ds_e2e.py --config usr/configs/midi/e2e/opencpop/ds100_adj_rel.yaml --exp_name $MY_DS_EXP_NAME`) needs same phoneme set size?

Issue - State: open - Opened by michaellin99999 about 2 years ago - 14 comments

#73 - custom phone_set file

Issue - State: open - Opened by michaellin99999 about 2 years ago - 1 comment

#72 - decoder part in e2e trainning using opencpop dataset

Issue - State: open - Opened by Liujingxiu23 about 2 years ago

#71 - I can't get checkpoint files.

Issue - State: closed - Opened by nakasako about 2 years ago - 6 comments

#70 - Migration to new code structures

Pull Request - State: closed - Opened by cyclekiller about 2 years ago

#69 - add the PLMS method

Pull Request - State: closed - Opened by luping-liu about 2 years ago

#68 - is there 22050 sample rate pretrain models

Issue - State: closed - Opened by 11721206 about 2 years ago - 1 comment

#67 - custom training data split

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

#66 - some questions about vocoder

Issue - State: open - Opened by LqNoob about 2 years ago

#63 - different spec_min and spec_max values

Issue - State: open - Opened by hubeibei007 over 2 years ago - 1 comment

#62 - mistake Run DiffSinger on Opencpop

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

#61 - fastspeech2推理问题

Issue - State: open - Opened by baipeng0110 over 2 years ago

#60 - MIDI SVS mode may produce uneven rhythms

Issue - State: open - Opened by yqzhishen over 2 years ago
Labels: enhancement, must-read

#59 - Inference SVS

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

#58 - Using higher sample rate for training and inference

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

#57 - ModuleNotFoundError with data preparation

Issue - State: closed - Opened by AkiKagura over 2 years ago - 1 comment
Labels: enhancement

#56 - I would like to receive the PopCS dataset

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

#55 - .wav files

Issue - State: closed - Opened by yangkang2318 over 2 years ago - 3 comments
Labels: solved

#54 - Missing file "popcs_statis.yaml"

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

#53 - help, ValueError: loaded state dict contains a parameter group that doesn't match the size of optimizer's group

Issue - State: closed - Opened by selyu504 over 2 years ago - 1 comment
Labels: Loading

#52 - Only using the pretrained model of HiFiGAN Singing

Issue - State: closed - Opened by powei-C over 2 years ago

#51 - Error: different shape of model parameter when generate example using diffsinger

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

#50 - help! data generate erros occured

Issue - State: closed - Opened by a1056632260 over 2 years ago - 2 comments
Labels: solved

#49 - 会不会出java版本？

Issue - State: open - Opened by watersoft123 over 2 years ago

#48 - Boundary Prediction

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

#47 - why set `uv = f0 = 0` in norm_interp_f0?

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

#46 - Pretraining FFT-Singer

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

#45 - Is there anyone covert model to torch script successfully?

Issue - State: open - Opened by kevink2424 over 2 years ago

#44 - train issue with private dataset

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

#43 - CUDA: out of memory

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

#42 - is the ph_dur from mel2ph or predicting by model？

Issue - State: closed - Opened by hertz-pj over 2 years ago

#41 - DiffSinger infer problem

Issue - State: closed - Opened by leon2milan over 2 years ago - 9 comments

#40 - Trian Opencpop data failed

Issue - State: closed - Opened by leon2milan over 2 years ago - 2 comments

#39 - I find that there are some bad cases in F0 prediction.

Issue - State: open - Opened by newportchen over 2 years ago - 7 comments

#38 - The phoneme duration label of POPCS Dataset.

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

#37 - multi speaker sing

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

#36 - How to do the DiffSinge test, why does my program report no errors but no audio files are generated?

Issue - State: open - Opened by baixinping618 over 2 years ago - 2 comments

#35 - Proposing a fix for inconsistent f0 length caused by different versions of parselmouth

Issue - State: closed - Opened by yuxinyuan over 2 years ago
Labels: enhancement

#34 - Some details about re-training MFA?

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

#33 - Looking forward for the midi data of POPCS~!

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

#32 - Data preparation Error at new commit 777d199

Issue - State: closed - Opened by zhangsanfeng86 over 2 years ago - 2 comments

#31 - Can you produce a version of MIDI with phoneme duration prediction

Issue - State: closed - Opened by newportchen over 2 years ago - 3 comments
Labels: enhancement, solved

#30 - Inference with unseen songs

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

#29 - How to inference on other <midi, text> files using the opencpop phone_set.json?

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

#28 - The model takes the phoneme duration as input when inference?

Issue - State: closed - Opened by YawYoung over 2 years ago - 5 comments
Labels: solved

#27 - 使用readme的预训练模型和步骤跑出来的语音不符合预期

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

#26 - How to train another language?

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

#25 - question about fs2 infer

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

#24 - Why feed in f0 in the midi version

Issue - State: closed - Opened by zhangsanfeng86 almost 3 years ago - 5 comments

#23 - The calculation of spec_min and spec_max

Issue - State: closed - Opened by Charlottecuc almost 3 years ago - 3 comments
Labels: enhancement, solved

#22 - Why it need to infer fs2 before training ds?

Issue - State: closed - Opened by hertz-pj almost 3 years ago - 2 comments
Labels: solved

#21 - How to convert the notes in 'opencpop/segments/transcriptions'?

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

#20 - What's the '0102_xiaoma_pe.zip file' for?

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

#19 - How to train HifiGAN-Singing from scratch?

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

#18 - How to train a model from scratch, with new data set?

Issue - State: closed - Opened by duheyu almost 3 years ago - 3 comments
Labels: solved

#17 - about 'An Easier Trick for Boundary Prediction'

Issue - State: open - Opened by keonlee9420 almost 3 years ago - 5 comments

#16 - AttributeError: 'HifiGAN' object has no attribute 'model'

Issue - State: closed - Opened by duheyu almost 3 years ago - 5 comments
Labels: solved

#15 - size mismatch for model.encoder_embed_tokens.weight: copying a param with shape torch.Size([62, 256]) from checkpoint, the shape in current model is torch.Size([57, 256]).

Issue - State: closed - Opened by ghost almost 3 years ago - 6 comments
Labels: solved

#14 - Data Preparation not working?

Issue - State: closed - Opened by ghost almost 3 years ago - 4 comments
Labels: solved

#13 - Share the hifi training code: task_cls: tasks.vocoder.hifigan.HifiGanTask

Issue - State: closed - Opened by zhangsanfeng86 almost 3 years ago - 2 comments
Labels: solved

#12 - When will release the code of PortaSpeech ?

Issue - State: closed - Opened by WhiteFu almost 3 years ago - 2 comments
Labels: solved

#11 - Possibly missing file between Step 1 and 2 for SVS?

Issue - State: closed - Opened by YangWangGit almost 3 years ago - 2 comments
Labels: solved

#10 - Question about Denoiser Residual Block

Issue - State: closed - Opened by tebin almost 3 years ago - 1 comment
Labels: solved

#9 - When will release the midi version

Issue - State: closed - Opened by zhangsanfeng86 almost 3 years ago - 2 comments
Labels: solved

#8 - please be patient when applying for datasets :)

Issue - State: open - Opened by MoonInTheRiver almost 3 years ago - 6 comments

#7 - Help to process the apply

Issue - State: closed - Opened by kevink2424 almost 3 years ago - 3 comments
Labels: solved

GitHub / MoonInTheRiver/DiffSinger issues and pull requests