DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to convert a vocal stem into your own singing voice with Resonant

Use MY VOICE and Seed-VC to turn an isolated vocal into a new WAV in your consented voice.

  • Content
  • Advanced
  • 120 minutes
  • From the README
  • Any agent

Voice cloning. Only convert a voice you own or have explicit permission to use, because Resonant requires that confirmation and records source and reference hashes.

Who it is forA singer or producer with a separated vocal stem who wants a performance sung in their own voice.

The problemRecording every take again is slow, and a full mix cannot be converted cleanly.

The outcomeA converted WAV on the Audio track that keeps the melody, timing, and phrasing of the source vocal.

What you need first

  • A Windows computer (the README describes Windows installers and a Windows workstation)
  • Optional: an NVIDIA GPU for ACE-Step and Seed-VC (ACE-Step recommends roughly 6 to 8 GB VRAM; Seed-VC needs an NVIDIA CUDA GPU with 6 GB VRAM minimum and 8 GB recommended)
  • Optional: an MCP-capable assistant such as Codex or Claude Code
  • For source builds only: Git for Windows and Node.js 22 or newer

Steps

  1. Install Resonant and open the header control labeled MY VOICE.
  2. Install and start the optional Seed-VC provider from inside Resonant.
  3. Choose an isolated source vocal and a clean 1 to 30 second reference recording you own.
  4. Confirm that you own the voice or have explicit permission to use it, then start the conversion.
  5. Check the converted WAV on the Audio track before arranging and mixing it.

Prompt for your agent

Using the Resonant repo at https://github.com/calesthio/Resonant, start a Seed-VC conversion of <source-vocal.wav> using the consented reference <my-voice.wav>, with both files inside the music workspace. Expected output: a converted WAV in the workspace; verify that voiceRightsConfirmed is true and that the conversion status reports complete before collecting the result.

Caveat. Seed-VC needs an NVIDIA CUDA GPU, uses about 5 GB after installer caches are cleaned, and is GPL-3.0 licensed and installed separately from Resonant.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for calesthio/Resonant and "Convert a vocal stem into your own singing voice"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer