santi-pdp/segan

Issues

• Unable to download the pretrained model.

  #98 opened 3 months ago by BINATHY
  0

• variable does not exist

  #80 opened 5 years ago by djjandXMU
  10

• no module named compat.v1

  #92 opened 2 years ago by aozorahime
  0

• noisy and clean wav files are the same in test_wav?

  #87 opened 4 years ago by LexiYIN
  0

• Could it be in real time ?

  #69 opened 5 years ago by ucasiggcas
  6

• Real-time

  #86 opened 4 years ago by msaad1311
  0

• SSNR is not as high as in the paper

  #62 opened 6 years ago by YinyanLiu
  7

• load failed when running train_segan.sh

  #51 opened 6 years ago by andreaschandra
  6

• OOM with small batch size on GPU

  #65 opened 5 years ago by Mega4alik
  1

• tensorflow.python.framework.errors_impl.NotFoundError: No registered 'TemporaryVariable' OpKernel for 'XLA_GPU' devices compatible with node {{node device_2/gradients/AddN_23/tmp_var}}

  #82 opened 4 years ago by zcy618
  1

• ValueError: Cannot feed value of shape (180149, 2) for Tensor 'Placeholder:0', which has shape '(180149,)'

  #84 opened 4 years ago by kurbobo
  1

• Could this model be called a real GAN? The discriminator might contribute nothing to the performance.

  #71 opened 5 years ago by ANYMS-A
  1

• why it takes too long

  #74 opened 5 years ago by NileZhou
  2

• Pre Trained Model

  #72 opened 5 years ago by predestination
  2

• ERROR when doing training

  #66 opened 5 years ago by YongyuG
  1

• My result is so bad, see my attachment

  #45 opened 7 years ago by papayazm
  1

• popd :not found

  #67 opened 5 years ago by ucasiggcas
  1

• Questions about voice pesq

  #78 opened 5 years ago by xiaojian10
  0

• Parameter size

  #76 opened 5 years ago by yunzqq
  0

• why my result used your SEGAN+ generator weights can't be a human voice?

  #75 opened 5 years ago by LittleFlyingSheep
  0

• weights

  #73 opened 5 years ago by soumimaiti
  0

• ValueError: Dimension 0 in both shapes must be equal, but are 1 and 16383. Shapes are [1] and [16383].

  #68 opened 5 years ago by fycheng-code
  5

• SEGAN with Wasserstein loss

  #50 opened 6 years ago by ahmed-fau
  1

• En error with Gradient disappearance

  #70 opened 5 years ago by marvin-nj
  0

• The program stuck at "Sampling some wavs to store sample references..."

  #44 opened 7 years ago by 867923756
  8

• dubug

  #64 opened 6 years ago by czh-learn
  0

• Model link not working

  #34 opened 6 years ago by subho406
  1

• discription of noise wavfiles

  #63 opened 6 years ago by ramesh720
  0

• How to resume training if interrupted

  #46 opened 7 years ago by imran7778
  4

• Error in using the pretrained model

  #61 opened 6 years ago by hdubey
  0

• InvalidArgumentError: Nan in summary histogram for: device_0/alpha_1

  #60 opened 6 years ago by uacusecha2000
  0

• some problems when load the model ane prediction

  #59 opened 6 years ago by fengqiyun
  0

• Loading model and prediction: error

  #43 opened 7 years ago by arijit17
  3

• Bad result when applied to "LibriSpeech" test-other wavs

  #58 opened 6 years ago by hoyden
  0

• Issue while testing audio

  #57 opened 6 years ago by HusainKapadia
  1

• Guideline on test audio files

  #38 opened 7 years ago by Mauker1
  8

• why cann't concat (x0 and diff) (tf版本问题，concat的参数位置换了)

  #56 opened 6 years ago by 353xiong
  0

• g_adv_loss

  #55 opened 6 years ago by jutn7537
  0

• for testing can we use smaller batch size?

  #47 opened 7 years ago by Benuraj
  1

• AttributeError: 'dict_values' object has no attribute 'op'

  #53 opened 6 years ago by maoxin7676
  1

• Can we cahnge the smapling frequency of audiowaves for training.

  #52 opened 6 years ago by imran7778
  6

• How much memory and time is needed to train?

  #36 opened 7 years ago by Mauker1
  4

• OutOfRangeError:RandomShuffleQueue‘’ is closed and has insufficient elements(required 150, current size 0)

  #49 opened 7 years ago by jutn7537
  5

• Parameters detail

  #48 opened 7 years ago by pig800509
  0

• aborted

  #40 opened 7 years ago
  1

• ResourceExhaustedError (see above for traceback): OOM

  #42 opened 7 years ago by mctyro
  1

• Fail for CuDNN 6.0

  #41 opened 7 years ago by mctyro
  1

• print('Sampling some wavs to store sample references...') # Hang onto a copy of wavs so we can feed the same one every time # we store samples to disk for hearing # pick a single batch

  #39 opened 7 years ago
  0

• Use 86 sentences to test this method but the result so bad.

  #37 opened 7 years ago by linan2
  3

• the model too big

  #33 opened 7 years ago by cfwang3
  0