Skip to main content

​
Migrating Voice Clones from ElevenLabs

If you have existing voice clones on ElevenLabs, you can batch-migrate them into Inworld using our open-source migration tool: github.com/inworld-ai/voice-migration-tool The tool runs locally on your machine and communicates directly with the ElevenLabs and Inworld APIs; it does not proxy your data through any additional intermediary servers.
Requires Node.js 18+ and ffmpeg installed on your machine.
1

Clone and start the tool

Run in your Terminal (macOS/Linux) or Command Prompt (Windows):
git clone https://github.com/inworld-ai/voice-migration-tool.git && cd voice-migration-tool && npm install && npm run dev
Then open http://localhost:3000 in your browser (or the localhost provided in Terminal).
2

Connect your accounts

Enter your ElevenLabs API key, Inworld API key, and Inworld workspace name. The tool will fetch all your user-created voice clones from ElevenLabs.
Only voices you created yourself are shown. ElevenLabs stock and professional voices are excluded.
3

Select voices and migrate

Select the voices you want to migrate, then click Migrate Selected. The tool downloads each voice’s audio samples from ElevenLabs and re-clones them into your Inworld workspace.Audio samples are automatically processed to meet Inworld’s requirements:
  • Converted to WAV format if needed
  • Padded to minimum 5 seconds if too short
  • Trimmed to 15 seconds if too long
4

Preview your migrated voices

After migration, click Preview on any voice to generate a sample utterance and hear it synthesized with Inworld TTS. This confirms the voice clone is working correctly in your workspace.