drethage/speech-denoising-wavenet

Issues

• requirements doesn't include TF

  #32 opened 6 years ago by Vichoko
  2

• Has someone do inference (denoise an audio) successfully?

  #44 opened 5 years ago by betegon
  2

• Denoised Speech is silence

  #26 opened 6 years ago by hermione0h
  3

• Implement the code

  #45 opened 3 years ago by 110318059Yu-Kai-Chen
  0

• Can't fine model

  #22 opened 6 years ago by ShagReza
  4

• out memory?

  #41 opened 5 years ago by yunzqq
  5

• Getting Out-of-memory (OOM) error on running the model on GPU.

  #9 opened 7 years ago by DillipKS
  5

• sor with shape[655360,31,1,256] and type float on /job:localhost/replica:0/task:0/device:CPU:0 by allocator cpu

  #43 opened 5 years ago by Wai-Yan-Phyo
  3

• Model just stops when denoising

  #21 opened 6 years ago by JSai23
  4

• Readme requirements

  #33 opened 6 years ago by Vichoko
  7

• Difference Between this and others

  #42 opened 5 years ago by DonaldTsang
  0

• parameter size

  #40 opened 5 years ago by yunzqq
  0

• Bad denoising results on the provided model

  #15 opened 7 years ago by kfmn
  2

• cudnn version

  #39 opened 5 years ago by LucySha
  0

• TypeError: 'float' object cannot be interpreted as an integer

  #34 opened 5 years ago by GrahamboJangles
  7

• Not using XLA:CPU for cluster because envvar TF_XLA_FLAGS=--tf_xla_cpu_global_jit was not set. If you want XLA:CPU, either set that envvar, or use experimental_jit_scope to enable XLA:CPU. To confirm that XLA is active, pass --vmodule=xla_compilation_cache=1 (as a proper command-line flag, not via TF_XLA_FLAGS) or set the envvar XLA_FLAGS=--xla_hlo_profile.

  #37 opened 5 years ago by GrahamboJangles
  0

• No version in python3 ?

  #36 opened 5 years ago by ucasiggcas
  1

• No optparse in python3

  #35 opened 5 years ago by ucas010
  0

• type error

  #18 opened 6 years ago by sdlfjsdlfkh
  1

• PESQ and STOI

  #31 opened 6 years ago by ffxz
  0

• Denoised audio 0db on NSDTSEA

  #28 opened 6 years ago by zfqy
  5

• Any way to prevent early stopping?

  #20 opened 6 years ago by butterl
  1

• index 2 is out of bounds for axis 0 with size 2

  #30 opened 6 years ago by MeiGM
  0

• Meaning of "in_memory_percentage" config parameter

  #29 opened 6 years ago by KDuzinkiewicz
  2

• Nvidia driver version exception

  #25 opened 6 years ago by gutsy-robot
  1

• What kind of HW is needed to run the "best performing model"?

  #27 opened 6 years ago by KDuzinkiewicz
  2

• About get_dataset(config, model) function

  #24 opened 6 years ago by zhengge
  0

• speed up training

  #1 opened 7 years ago
  3

• Trying to reproduce the results/ Error with pygpu

  #23 opened 6 years ago by AkhilSinghRana
  0

• ValueError: output of generator should be a tuple (x, y, sample_weight) or (x, y). Found: None

  #12 opened 7 years ago by DillipKS
  3

• How you got the best performing model

  #16 opened 7 years ago by shifaspv
  0

• Denoised audio has 0db?

  #13 opened 7 years ago by wuweijia1994
  3

• How to predict not a single value but a distribution as in original Wavenet?

  #14 opened 7 years ago by kfmn
  3

• Any other audio format supported by Wavenet model (like .mp3) other than .wav?

  #11 opened 7 years ago by DillipKS
  2

• Why does the trained model need clean audio files for inferencing on the same noisy audio files?

  #10 opened 7 years ago by DillipKS
  3

• Value error

  #8 opened 7 years ago by kiranbomanchi
  0

• got wrong array shape

  #6 opened 7 years ago by tonyyyye
  3

• Where is the pre-trained model?

  #7 opened 7 years ago by BernardoAflalo
  1

• Is there any possibilities to change into Tensorflow model?

  #4 opened 7 years ago by tonyyyye
  1

• Issue on using the trained model

  #5 opened 7 years ago by shifaspv
  6

• Method to evaluate enhanced speech

  #3 opened 7 years ago by lifelongeek
  2

• Some samples of the denoised speech wave is missing in the end of the speech

  #2 opened 7 years ago by wyn314
  1