Skip to content

issues Search Results · repo:zcaceres/spec_augment language:"Jupyter Notebook"

Filter by

15 results
 (107 ms)

15 results

inzcaceres/spec_augment (press backspace or delete to remove)

Hi, I have noticed an issue with your time warping and it s already mentioned in #12. I think that not how time warp should be (maybe my opinion is wrong since I m not familiar with TF so I can t try tfa.image.sparse_image_warp ...
  • vlozg
  • 2
  • Opened 
    on Jul 5, 2021
  • #17

`def time_warp(spec, W=5): num_rows = spec.shape[1] spec_len = spec.shape[2] device = spec.device y = num_rows//2 horizontal_line_at_ctr = spec[0][y] assert len(horizontal_line_at_ctr) == spec_len point_to_warp ...
  • PrathamSoni
  • Opened 
    on Jan 22, 2021
  • #15

Have I misunderstood that SpecAugment could be used to convert speech to text? It seems from your Medium article that it is simply something that enhances the process of one training a model rather than ...
  • youssefavx
  • 1
  • Opened 
    on Sep 3, 2020
  • #14

Hi, I am trying to run your code install.sh file in conda environment and getting an error cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++ In file included ...
  • Aditya3107
  • 1
  • Opened 
    on Aug 26, 2020
  • #13

Hi, I have a question whether the time warping can sometimes cause a spectrogram to be too distorted or not. Screen Shot 2563-07-22 at 14 37 29 I applied Time Warping with W=2 and it sometimes results ...
  • tann9949
  • 10
  • Opened 
    on Jul 22, 2020
  • #12

Hi, I found some code in nb_SparseImageWarp.py need changes in order to work in Pytorch 1.4.0: In line 115: X, LU = torch.gesv(rhs, lhs) - X, LU = torch.solve(rhs, lhs) (torch.gesv has been deprecated) ...
  • lukewys
  • 3
  • Opened 
    on Apr 20, 2020
  • #11

Hi, is there any thing I can do to convert the mel spectrogram back to audio? I am using librosa to covert it but it takes ages just to reform 8-10s audio. Thanks
  • junaedifahmi
  • 1
  • Opened 
    on Dec 12, 2019
  • #10

It seems that the shape of input image is (b, h, w), while the same function in TensorFlow can deal with color image. Is the library able to warp color image(3 channel) by simply add a dismension on code? ...
  • wtjiang98
  • 2
  • Opened 
    on Aug 3, 2019
  • #9

Cloning into torchaudio ... remote: Enumerating objects: 87, done. remote: Counting objects: 100% (87/87), done. remote: Compressing objects: 100% (62/62), done. remote: Total 1080 (delta 47), reused ...
  • hamilyon
  • 3
  • Opened 
    on Jul 16, 2019
  • #8

In my initial experiments I got some promising results with our ASR library. Can I include a modified py version in our library? How can I reference your work? I didn t see any license anywhere.
  • bobchennan
  • 1
  • Opened 
    on May 13, 2019
  • #7
Issue origami icon

Learn how you can use GitHub Issues to plan and track your work.

Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub Issues
ProTip! 
Restrict your search to the title by using the in:title qualifier.
Issue origami icon

Learn how you can use GitHub Issues to plan and track your work.

Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub Issues
ProTip! 
Press the
/
key to activate the search input again and adjust your query.
Issue search results · GitHub