facebookarchive/loop

Issues

• Block on preprocessing

  #78 opened 5 years ago by zhang-yunke
  0

• hello world text

  #77 opened 5 years ago by captainsan
  0

• TBPTTIter.split_length() error

  #75 opened 5 years ago by hash2430
  1

• Batch

  #76 opened 5 years ago by hash2430
  0

• Strange fail on "The quick brown fox jumps over the lazy dog."

  #74 opened 5 years ago by mrgloom
  0

• Look like it fails on '!' character.

  #73 opened 5 years ago by mrgloom
  0

• ERROR: Failed to find norm file.

  #69 opened 5 years ago by mrgloom
  7

• Understanding feat tensor dimensions

  #72 opened 5 years ago by mrgloom
  1

• How this repo compared to Merlin?

  #71 opened 5 years ago by mrgloom
  0

• ImportError: No module named torch

  #70 opened 5 years ago by mrgloom
  0

• bash scripts/download_tools.sh failed on Mac OS

  #68 opened 5 years ago by mrgloom
  0

• Parameters for dataset in the wild

  #58 opened 6 years ago by aomv
  12

• Train VCTK dataset for all speakers

  #66 opened 5 years ago by SibtainRazaJamali
  0

• Error when 'make' HTK-3.4.1 and hts_core.

  #65 opened 5 years ago by hatiff
  0

• Error running train.py

  #63 opened 5 years ago by sharonzhou
  1

• No matching distribution found for phonemizer (from -r scripts/requirements.txt (line 5))

  #61 opened 5 years ago by ethanjyx
  1

• Any clue? Missing norm_info_mgc_lf0_vuv_bap_63_MVN.dat while prepossessing Lj Speech Dataset.

  #48 opened 6 years ago by niravpatel2008
  7

• Issue running install_tts.py to preprocess data

  #62 opened 6 years ago by wanshun123
  1

• Main Readme wav files are missing and first instruction doesn't work

  #59 opened 6 years ago by ryancwalsh
  1

• Using pre-trained model for new speaker?

  #60 opened 6 years ago by xerdink
  0

• Trump voice

  #54 opened 6 years ago by aomv
  1

• Out of memory in validation step

  #57 opened 6 years ago by sausax
  1

• Error while generating sample speech from pretrained model

  #56 opened 6 years ago by ajinkyakulkarni14
  2

• How much minutes of audio datasets to train for a single speaker using blizzard model?

  #28 opened 7 years ago by jaxlinksync
  47

• Recreating blizzard baseline

  #55 opened 6 years ago by ghostcow
  0

• accuracy of the model by using vctk

  #53 opened 6 years ago by vinaykumarbundla
  0

• Training Error. Epoch 5.

  #50 opened 6 years ago by PetrochukM
  3

• visualize flag not working with visdom

  #52 opened 6 years ago by ghostcow
  1

• What does a good loss curve look like?

  #51 opened 6 years ago by PetrochukM
  2

• How much hours or minutes of training data do we need to make the performance very good?

  #45 opened 6 years ago by jaxlinksync
  1

• Creating norm_info_mgc_lf0_vuv_bap_63_MVN.dat for the Full VCTK dataset

  #49 opened 6 years ago by PetrochukM
  7

• How to train the samples obtained from public？too much noise and speed too fast!

  #37 opened 6 years ago by lvenoxi
  1

• Pretrained checkpoints for Blizzard 2013 and LJSpeech

  #46 opened 6 years ago by PetrochukM
  1

• Full VCTK dataset

  #47 opened 6 years ago by PetrochukM
  1

• Blizzard Model

  #44 opened 6 years ago by PetrochukM
  6

• From English into Chinese?

  #38 opened 6 years ago by ChrisChan2013
  1

• espeak or espeak-NG?

  #43 opened 6 years ago by PetrochukM
  2

• Why scale the outputs by 30 in Decoder.update_buffer?

  #40 opened 6 years ago by arvieFrydenlund
  1

• Reproducing the results

  #41 opened 6 years ago by pfriesch
  9

• Error during training

  #39 opened 6 years ago by jaxlinksync
  2

• Confusion of the Phoneme Generation

  #42 opened 6 years ago by darongliu
  1

• What is the split ratio between the training set and test set of the VCTK data provided in your project?

  #32 opened 7 years ago by jaxlinksync
  2

• Training on big dataset.

  #35 opened 7 years ago by hepower
  1

• Can loop support parallel training in multiple GPU?

  #34 opened 7 years ago by hepower
  1

• ' out of memory' accrued when training with single speaker

  #36 opened 7 years ago by hepower
  2

• Face 'Not a finite gradient or too big, ignoring.' when training other data.

  #33 opened 7 years ago by hepower
  1

• Can we use multi-GPU to make the training faster? How?

  #30 opened 7 years ago by jaxlinksync
  2

• Error when synthesis wave with the download model.

  #31 opened 7 years ago by hepower
  1

• Does anyone successfully train a multi-speaker loop model with his/her own data?

  #29 opened 7 years ago by hepower
  0

• Training Epochs Number

  #27 opened 7 years ago by gaziway
  1