NVIDIA/mellotron

Issues

• What is the reason of filtering "_" and "~" symbols?

  #125 opened a year ago by kremnik
  0

• Here's some code to start mellotron inference by calling a .py file from CLI [Docs]

  #121 opened 2 years ago by mepc36
  0

• Synthesize own text without style transfer gives poor audio results

  #120 opened 2 years ago by ocesp98
  1

• How to generate .musicxml files like the examples in `/data`?

  #119 opened 2 years ago by mepc36
  1

• colab error

  #98 opened 4 years ago by mochacat2001
  2

• Two key points of training multispeaker mellotron

  #112 opened 3 years ago by WelkinYang
  0

• colab demo for inferenece

  #114 opened 3 years ago by loboere
  0

• how to train?

  #113 opened 3 years ago by loboere
  0

• Training time

  #110 opened 3 years ago by Rongjiehuang
  0

• inference speed on CPU

  #109 opened 3 years ago by Adibian
  0

• Try to train some new words

  #108 opened 3 years ago by zfishbone01
  0

• Voice synthesis by model is not the same as the voice with speaker ID

  #107 opened 3 years ago by tuanh123789
  1

• number of epoch issue

  #81 opened 4 years ago by yuzuda283
  1

• Training on EmovDB

  #105 opened 3 years ago by kngan43
  2

• Training on a different language

  #102 opened 3 years ago by OSSome01
  1

• parse_output error with Blizzard2013 data

  #104 opened 3 years ago by jinhonglu
  0

• error in the model training on libritts data

  #76 opened 4 years ago by raymond00000
  9

• Inference without rhythm and pitch

  #103 opened 3 years ago by kngan43
  0

• Mismatch model volume

  #101 opened 3 years ago by thkaang
  0

• NoneType' object is not iterable

  #100 opened 4 years ago by shaokun11
  0

• Can I use TensorRT to speed up model inference?

  #97 opened 4 years ago by doudoucao
  0

• Something wrong with text padding

  #96 opened 4 years ago by arijitx
  5

• Is there a way to extract character durations?

  #95 opened 4 years ago by jmasterx
  0

• hparams for libritts training

  #92 opened 4 years ago by krishnashankar
  0

• Could you tell me why use speaker id?

  #93 opened 4 years ago by Moon-sung-woo
  0

• Where is the singer datdaset?

  #91 opened 4 years ago by c1a1o1
  1

• Help needed w.r.t. inference

  #90 opened 4 years ago by nirmeshshah
  0

• Can I transfer only rhythm or pitch between two audio ？

  #89 opened 4 years ago by niu0717
  0

• How can I speed up or slow down the output audio?

  #86 opened 4 years ago by BabaiLi
  0

• rhythm not good so the inference audio output is weird？

  #88 opened 4 years ago by ryanjfdeng1
  0

• Inference error

  #85 opened 4 years ago by BabaiLi
  1

• why ignore speaker embedding?

  #72 opened 4 years ago by zzw922cn
  2

• How to choose a target speaker for generating voice

  #74 opened 4 years ago by deepuvikraman
  5

• TypeError: create_target_machine() got an unexpected keyword argument 'jitdebug'

  #80 opened 4 years ago by zxwwxz001
  0

• Inference with our own text on pretrained model

  #78 opened 4 years ago by xw1324832579
  0

• Strange query input shape of multihead attention in STL

  #77 opened 4 years ago by elch10
  1

• Inference failed

  #66 opened 5 years ago by lqniunjunlper
  1

• Style not being applied

  #64 opened 5 years ago by kannadaraj
  4

• [QUESTION] About lyrics hyphenation

  #58 opened 4 years ago by loretoparisi
  5

• Adding another speaker

  #61 opened 4 years ago by JakubReha
  5

• Difference between CMUDict of None?

  #67 opened 5 years ago by lqniunjunlper
  1

• Inference bug?

  #65 opened 4 years ago by lqniunjunlper
  1

• require large memory

  #69 opened 4 years ago by aijianiula0601
  3

• Question about the gender of the Libri dataset used by Mellotron mention in the published article

  #70 opened 4 years ago by TitiAffandi
  1

• ReferenceEncoder did not use the actual mel lengths

  #71 opened 4 years ago by hubeibei007
  2

• Gradient overflow with Mixed Precision Training

  #63 opened 5 years ago by MinHyung-Kang
  1

• Speaker order is randomised while loading - why?

  #68 opened 5 years ago by karkirowle
  1

• The speaker ids are misaligned in inference.ipynb

  #59 opened 5 years ago by pneumoman
  7

• Using own text to generate speech using Mellotron

  #62 opened 5 years ago by astricks
  8

• Question about custom dataset

  #60 opened 5 years ago by LucasRotsen
  2