CastPolish: Free, Open-Source Audio Processing for Your Podcast

CastPolish is a free, open-source Mac app that brings professional audio post-production to your podcast — without any cloud service, subscription fees, or your audio ever leaving your computer. It handles loudness normalization, noise reduction, transcription, AI-generated chapter titles, summaries, and shownotes, all running locally on your Mac.

What CastPolish Does

• Audio processing — two-pass EBU R128 loudness normalization to −16 LUFS (broadcast podcast standard), highpass filter, compression, and peak limiting
• Optional noise reduction — ffmpeg's built-in afftdn FFT noise filter (checkbox, adds ~20 seconds)
• Transcription — faster-whisper with word-level timestamps, exported as HTML with audio player, WebVTT captions, and JSON
• AI shownotes — chapter titles, long and brief summaries, and suggested tags via Ollama running locally (no API key needed)
• Speaker diarization — optional, via pyannote.audio (requires a free HuggingFace account)
• macOS app — a double-click .app launcher with Dock icon that starts the server and opens your browser automatically

Get It on GitHub

CastPolish is fully open-source under the MIT license. Source code, issue tracker, and documentation are at github.com/abc3-Mac/castpolish.

Installing CastPolish on Your Mac

Open Terminal by pressing Command + Space, typing Terminal, and pressing Return. Paste each command below one at a time, pressing Return after each one and waiting for it to finish before moving to the next.

Step 1 — Install Homebrew

Homebrew is a package manager that installs software on your Mac:

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

It will ask for your Mac password (you won't see the characters as you type). If it shows a message saying "Run these two commands" when finished, paste those two lines and press Return after each.

Step 2 — Install Python and ffmpeg

brew install python ffmpeg

Step 3 — Install Ollama (optional — for AI summaries)

brew install ollama
ollama pull llama3.2

The second command downloads an AI model (~2 GB). Skip this step if you don't need AI summaries — CastPolish works fine without it.

Step 4 — Download CastPolish

git clone https://github.com/abc3-Mac/castpolish.git
cd castpolish

Step 5 — Install dependencies

python3 castpolish.py setup

Step 6 — Build the Mac app

python3 create_macos_app.py

When it asks "Open folder in Finder? [Y/n]" — type Y and press Return. A Finder window opens showing CastPolish.app. Drag it to your Applications folder or Dock.

You're Done!

Double-click CastPolish.app to launch it. Your browser opens automatically at http://localhost:8765. The first time you process a file it downloads the Whisper transcription model (~500 MB) — this only happens once.