Generated audio is noisy and jittered when running via Gradio

Open
#228 4 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
3/5
Estimated time
1-2 days
Newbie friendliness
38/100
Issue type
Bug
Clarity
Needs clarification
Activity status
Quiet
Tech stack
python, pytorch

Research direction

No source files or tests are named. First reproduce Stable Audio through the Gradio interface with torch 2.5.1, torchaudio 2.9.0, and torchcodec 0.8.1, testing both prompted and unprompted generation; trace the generation path to identify where the noisy, jittered output is introduced. Done means both cases produce clean audio and a regression check covers the failure.

Written by the indexing model from the issue text.

Description

Hello, and thank you for your amazing work on making this fully open-source!

While testing Stable Audio through Gradio, I encountered an issue: every time I run the model — both with a prompt and without one — the generated audio turns out extremely noisy and jittered.

Do you have any idea what might be causing this issue, or any suggestions for how to fix it?
I’d really appreciate any guidance.

Thanks again for your great work!

Image Image I'm using torch 2.5.1 with torchaudio 2.9.0, torchcodec 0.8.1
Dominant language
Python
Stars
3.9k
Forks
490
PR merge metrics
No merged PRs in 30d

Getting set up

This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from Stability-AI/stable-audio-tools

All issues in Stability-AI/stable-audio-tools

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.