Insights: pytorch/audio

1 Pull request merged by 1 person

• Experimenting with Pytorch 2.0 compile() optimizations on wav2vec2 models

  #2941 merged Dec 27, 2022

5 Pull requests opened by 4 people

• Use masked input for negative sampler

  #2933 opened Dec 21, 2022

• Add subtractive synthesis tutorial

  #2934 opened Dec 22, 2022

• Add preemphasis and deemphasis transforms

  #2935 opened Dec 22, 2022

• Fix confusing wording in the documentation

  #2937 opened Dec 24, 2022

• Refactor a part of convert_image

  #2940 opened Dec 27, 2022

4 Issues closed by 1 person

• RuntimeError: torchaudio.sox_effects.sox_effects.apply_effects_tensor requires sox

  #2938 closed Dec 25, 2022

• Limit StreamReader to use a specifed number of cores

  #2855 closed Dec 25, 2022

• StreamReader requires the start of the next frame before returning the decoded previous frame

  #2887 closed Dec 25, 2022

• [Migration] Torchaudio Complex Tensor Support and Migration

  #1337 closed Dec 22, 2022

1 Issue opened by 1 person

• HTDemucs (Hybrid Transformer Demucs)

  #2936 opened Dec 23, 2022

5 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Add HiFiGAN bundle

  #2921 commented on Dec 22, 2022 • 4 new comments

• DDSP components in TorchAudio

  #2835 commented on Dec 22, 2022 • 0 new comments

• Nightly build [DO NOT DELETE OR MERGE]

  #306 commented on Dec 27, 2022 • 0 new comments

• [WIP] Add DSP ops

  #2808 commented on Dec 22, 2022 • 0 new comments

• Refactor cpp modules

  #2930 commented on Dec 27, 2022 • 0 new comments