jishengpeng/WavTokenizer

Issues

• Training time

  #63 opened 21 days ago by Ming-er
  1

• Question about the validation set

  #62 opened a month ago by alberthli
  6

• The loss value when the model converges

  #20 opened 4 months ago by yangyyt
  10

• 重建效果是否符合预期

  #61 opened a month ago by zzchust
  3

• Performance in LLM-based-TTS

  #40 opened 4 months ago by Liujingxiu23
  4

• # audio_tokens错误问题

  #60 opened a month ago by EdisonZhu33
  1

• What is the expected behaviour with changing the bandwidth parameter

  #51 opened 2 months ago by tanmaylaud
  1

• We update WavTokenizer paper in Arxiv and release WavTokenizer-Large checkpoint in Huggingface on 2024.10.22

  #45 opened 3 months ago by jishengpeng
  5

• how's the quality compar e with other audio tokenizer?

  #59 opened 2 months ago by MonolithFoundation
  3

• Duplicate discriminator in dac?

  #58 opened 2 months ago by npuichigo
  4

• question about streaming infer

  #56 opened 2 months ago by VJJJJJJ1
  2

• Usage for speech separation and temporal audio features

  #29 opened 4 months ago by saveriyo
  7

• some questions about model

  #49 opened 2 months ago by VJJJJJJ1
  1

• Traning on wenetspeech couldn‘t converge

  #28 opened 4 months ago by dyyoungg
  3

• worse performance of large model compared to small model?

  #54 opened 2 months ago by XiaoshanHsj
  4

• Semantic Representation

  #55 opened 2 months ago by Uneasy-Z
  2

• Model can not converge

  #52 opened 2 months ago by VJJJJJJ1
  2

• Training for wav to midi transcriper

  #53 opened 2 months ago by mito0o852
  1

• Alignment language vocabulary and speech space

  #22 opened 4 months ago by varfolomeeff
  3

• Questions for Creating a Better Model

  #50 opened 2 months ago by ootsuka-repos
  2

• Streaming infer

  #44 opened 3 months ago by wntg
  3

• Installable Package

  #48 opened 2 months ago by poonehmousavi
  2

• Config/Model Checkpoint Pairing

  #47 opened 3 months ago by MorenoLaQuatra
  5

• Question about Audio Preprocessing

  #46 opened 3 months ago by xjf-303
  1

• how to train the model with Token/s about 23, that is hopsize=1024

  #35 opened 4 months ago by Liujingxiu23
  5

• probability density for each index in the codebook

  #41 opened 3 months ago by goforher
  1

• How many training steps to train wavtokenizer?

  #43 opened 3 months ago by sphmel
  1

• When will the large unify model (speech, music, audio) be released?

  #42 opened 3 months ago by MrPig
  1

• Files Missing？

  #39 opened 4 months ago by goforher
  6

• CER Performance of Reconstructed Audio

  #34 opened 4 months ago by howitry
  6

• why grad norm is so high？

  #38 opened 4 months ago by necrophagists
  1

• Why so large commit loss weight

  #36 opened 4 months ago by Ming-er
  4

• speech medium v2

  #37 opened 4 months ago by theodorblackbird
  1

• Question about training

  #32 opened 4 months ago by handsomelys
  2

• Using EMA on the generator markedly improves the validation loss

  #33 opened 4 months ago by erogol
  2

• Maximum duration supported during inference?

  #31 opened 4 months ago by LiuShixing
  1

• encounter shape inconsistent in training 16kHz

  #19 opened 4 months ago by dyyoungg
  3

• How many hours of Chinese data are there?

  #30 opened 4 months ago by LiuShixing
  1

• Comparison with Whisper

  #27 opened 4 months ago by isruihu
  1

• What is the difference between the config for training WavTokenizer-small and WavTokenizer-large?

  #25 opened 4 months ago by handsomelys
  2

• WavTokenizer-mdium is release on 2024.09.09

  #23 opened 4 months ago by jishengpeng
  4

• Future 48kHz model

  #21 opened 4 months ago by Ronsor
  1

• Purpose of os.environ['CUDA_LAUNCH_BLOCKING'] = '1' in train.py

  #17 opened 4 months ago by seastar105
  1

• MRD vs MS-STFTD

  #10 opened 4 months ago by Yagelmx
  4

• Mel or wav？

  #18 opened 4 months ago by howitry
  1

• fail to install

  #12 opened 4 months ago by JoyceMind
  2

• Weight of model

  #16 opened 4 months ago by JoyceMind
  1

• Please consider about 16K model?

  #15 opened 4 months ago by ywh-my
  1

• About infer in GPU

  #13 opened 4 months ago by JohnFengNeumann
  1

• encode and decode for "16k sample"

  #8 opened 4 months ago by sunnnnnnnny
  1