elevenlabs-voice-isolator

Name: elevenlabs-voice-isolator
Availability: InStock
Author: inference.sh

Removes background noise and isolates vocals from audio files via the inference.sh CLI.

723|101|Updated Jan 31, 2026

Search Tags:#podcast #ElevenLabs #denoise #audio cleanup #noise removal #vocal extraction #voice isolation

Ready to run

npx skills add https://github.com/inference-sh/skills --skill elevenlabs-voice-isolator

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: elevenlabs-voice-isolator
Source: https://github.com/inference-sh/skills/tree/main/tools/audio/elevenlabs-voice-isolator
Command: npx skills add https://github.com/inference-sh/skills --skill elevenlabs-voice-isolator

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Noisy recordings with ambient sound, background music, or echo degrade podcasts, interviews, and voiceovers, and manual audio cleanup requires specialized editing tools and expertise.

Core Features & Use Cases

Noise Removal: Strips ambient noise, background music, reverb, wind, traffic, and electrical hum from audio recordings.
Vocal Isolation: Extracts clean voice tracks from mixed audio, supporting WAV, MP3, FLAC, OGG, and AAC up to 500MB and 1 hour.
Pipeline Integration: Chains with transcription, voice changing, and video merging workflows through the belt CLI.
Use Case: A podcaster records an interview in a café; run the voice isolator to remove background chatter, then transcribe the cleaned audio with speaker diarization.

Quick Start

Ask the agent to isolate the voice and remove background noise from your audio file URL using the ElevenLabs voice isolator.

Frequently Asked Questions about elevenlabs-voice-isolator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I remove background noise from an audio recording?▼

Run the ElevenLabs voice isolator through the belt CLI with your audio URL as input: belt app run elevenlabs/voice-isolator with the audio parameter. It returns a cleaned audio file with ambient noise, music, and echo removed.

What audio formats does ElevenLabs voice isolator support?▼

The voice isolator supports WAV, MP3, FLAC, OGG, and AAC formats. Each file can be up to 500MB in size and up to one hour in duration.

Can I extract vocals from a song for a remix or karaoke?▼

Yes, the voice isolator separates vocals from instrumental tracks. Pass the song URL as the audio input and it returns the isolated vocal track, which works for remixes, karaoke, or acapella extraction.

How do I transcribe a noisy recording accurately?▼

First clean the audio with the voice isolator, then transcribe the result with ElevenLabs STT using the diarize option for speaker separation. Cleaning first improves transcription accuracy on noisy sources.

What are the limitations of AI voice isolation?▼

Files are limited to 500MB and one hour, and the tool targets non-voice sounds like noise, music, and reverb. Overlapping speakers or extremely degraded sources may still retain artifacts after processing.