DEDUPED Hot AI repos, turned into use cases Get 5 a day

How to remove noise and reverb from a voice recording with AuK

Restore a noisy or reverberant speech clip to clearer audio with one instruction and no separate tools.

  • Content
  • Intermediate
  • 1 hour
  • From the README
  • Any agent

Authorization notice. AuK is dual-use. Only use it on software, sites and systems you own or have explicit permission to test.

Who it is forA researcher or creator with field recordings or home-studio takes that have background noise or echo.

The problemNoisy or echoey recordings are hard to use in a finished project without manual cleanup.

The outcomeA cleaned WAV file of the speech, with the same length as the input.

What you need first

  • Python 3.10 and uv or Conda
  • A GPU with enough VRAM to run the model (CPU offload is CUDA only)
  • Disk space and network access to download the AuK weights and the Qwen2.5-Omni-3B encoder from Hugging Face or ModelScope
  • Source audio or a reference clip for tasks that need one

Steps

  1. uv venv --python 3.10
  2. source .venv/bin/activate
  3. uv pip install -e .
  4. hf download tencent/AuK --local-dir ./ckpts/AuK
  5. hf download Qwen/Qwen2.5-Omni-3B --local-dir ./ckpts/Qwen2.5-Omni-3B (inferred, check the README)
  6. auk-infer --audio <noisy.wav> --instruction "Restore this audio to clean speech: keep all speakers, remove noise and reverberation, and keep the same length." --output <clean.wav> (inferred, check the README)

Prompt for your agent

In the Tencent-Hunyuan/AuK repo, run auk-infer on <noisy.wav> with an instruction asking for noise and reverberation removal, and save the output as <clean.wav>. Verify that <clean.wav> exists and has the same duration as <noisy.wav>.

Caveat. The README's enhancement example uses a Chinese instruction, so an English instruction is untested in the README and may need tuning.

Share: X LinkedIn Hacker News Email Copy link

Disclaimer

Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page