AudioEditor

To remove filler words and dead air from a podcast recording, this Skill uses Whisper for word-level transcription and Claude for analysis to identify precise cuts. It then automatically removes stutters and false starts, applying crossfades for seamless audio cleanup.

The best way to automate audio cleanup for video editing is using AI-powered analysis to detect and cut unwanted segments. This Skill transcribes the audio, intelligently identifies dead air and filler words, and executes automated edits with crossfades via ffmpeg.

Yes, automated audio editing works with MP3 files by processing the input through Whisper for transcription and ffmpeg for cutting. You can directly clean audio files like 'my_recording.mp3' to remove filler words and fix dead air automatically.

Yes, you can use the Cleanvoice API for optional cloud polishing after removing filler words. This integration performs advanced mouth sound removal and audio normalization, resulting in a professional-sounding final podcast product.

Intelligent cut detection for fixing audio works by generating a word-level transcript via Whisper, then analyzing the text to identify stutters, false starts, and long pauses. It automatically applies precise cuts and crossfades to clean the dead air.

You do not need an API key to perform standard podcast cleanup, as transcription and editing use Whisper, Claude, and ffmpeg locally. An API key is only required if you choose the optional cloud polish feature via the Cleanvoice API.