From zero to first extraction in 5 minutes.

Everything you need to download, install, and run your first extraction.

1

Download

Grab the latest version of XTC Studio and open the installer.

Download for macOS
macOS 13+ Ventura or later
8 GB RAM Minimum required
2 GB disk space For dependencies

XTC Studio is a native macOS app. Windows and Linux support is planned.

2

Install

Open the app. XTC Studio will check for dependencies and install them automatically.

Here is what the auto-installer sets up for you:

  • Homebrew (if not already present)
  • Conda environment
  • Python 3.10+
  • Whisper models
  • yt-dlp
Checking Homebrew... ✓ installed Creating conda env... ✓ xtcstudio-env Installing Python 3.10.. ✓ complete Downloading Whisper... ✓ base model ready Installing yt-dlp... ✓ ready

First launch takes 2-3 minutes. After that, the app starts instantly.

3

API Keys

To use AI extraction, add at least one LLM API key.

Claude (Anthropic)

console.anthropic.com

Recommended for most extractions.

OpenAI

platform.openai.com

For GPT-4o access.

Grok (xAI)

console.x.ai

Alternative option.

Paste your key in Settings. Keys are stored locally, never sent to XTC servers.

Transcription works without an API key. Only AI extraction requires one.

4

First Extraction

Follow these steps to run your first extraction from start to finish.

  1. Go to the Extract tab
  2. Select YouTube (or any supported platform)
  3. Paste a video URL
  4. Click "Download" to fetch the audio
  5. Switch to the Transcribe tab, pick an engine (start with Fast Draft)
  6. Click "Transcribe Selected"
  7. Enable AI, choose Claude > Sonnet
  8. Select a mode (Quotes is a great start)
  9. Run extraction
Video: Step-by-step walkthrough of your first extraction
Screenshot: extraction results
5

Organize and Export

After extraction, switch to the Organize tab to cluster results by theme.

Use the Select tab to curate: keep the best extractions, discard the rest.

Export to your preferred format from the Publish tab.

.md .json .txt .pdf .docx .csv

Full Video Walkthrough

Video: Complete zero-to-extraction walkthrough

Tips and Tricks

Frequently Asked Questions

Do I need an internet connection?

For downloading content and AI extraction, yes. Transcription with Whisper works offline after the first model download.

Is my data stored on your servers?

No. Everything runs locally on your Mac. Your transcripts, extractions, and API keys never leave your machine.

Which transcription engine should I use?

Start with Fast Draft for quick results. Use Studio Pro when you need speaker labels and higher accuracy. Use AssemblyAI for the best quality on critical content.

Can I process multiple videos at once?

Yes. Paste multiple URLs in the Extract tab, one per line. You can also import entire YouTube channels or playlists.

What languages are supported?

Whisper supports 90+ languages. AI extraction quality depends on the LLM, but Claude and GPT handle most major languages well.

How do I update XTC Studio?

The app checks for updates on launch. You can also re-download the latest version from this page.

Need help? Join the community.

Get answers, share workflows, and connect with other creators.

Join the Community

Stay in the loop

Get notified about new features and updates.