Caption Comet
Local AI Video Captions
Only for Mac
Free · In‑App Purchases
Mac
Local AI transcription with dual-model review, automatic hallucination correction, powerful caption editing, and no subscription.
Automatically transcribe, review, edit, and export captions using private AI that runs locally on your Mac.
Caption Comet automatically detects the spoken language, corrects likely transcription hallucinations in the background, and gives you precise control over text, timing, and appearance. Your media stays on your Mac: no account, API key, cloud transcription service, or subscription required.
DUAL-MODEL REVIEW
For supported languages, Caption Comet independently transcribes your media using two different AI models, compares the results, and shows you where they disagree.
Instead of blindly proofreading every caption, jump straight to the places most likely to need your attention.
Dual-model review is available in Project Mode and for individual files using Review Before Export in Quick Mode.
STUDY MODE IS ALWAYS FREE
Batch transcribe audio, video, and entire folders with no 30-second limit. Combine folder transcripts for easy studying or summarization, or extract text from existing subtitle tracks.
Study Mode uses study-friendly timestamps rather than precise subtitle timing and is designed for lectures, classes, interviews, research, and other material you want to read, search, or summarize.
No Pro purchase is required.
MAKE CAPTIONS YOUR OWN
• Traditional, karaoke, and word-by-word captions
• Custom fonts, colors, sizes, outlines, shadows, and backgrounds
• Drag captions exactly where you want them
• Add watermarks and QR codes
• Save reusable styles as templates
• Burn captions directly into finished video
EDIT CAPTIONS WORD BY WORD
Correct individual words, adjust timing, split and merge phrases, preview changes with your video, and shift generated captions earlier or later with precise Word Timing Offset controls.
QUICK MODE AND PROJECT MODE
Quick Mode makes everyday jobs simple: choose reusable output settings, drop in files or folders, and let Caption Comet automatically detect the language, transcribe, and export. Batch jobs can mix resolutions, aspect ratios, containers, and codecs.
Project Mode provides the complete editing and review workflow for videos that need more attention. You can also batch files into editable projects and review them individually.
SIMPLE MODEL SETUP
Whisper Large V3 provides broad language support. Two optional specialist models expand dual-model review:
• Parakeet TDT 0.6B V3 for English and European languages
• Qwen3-ASR 1.7B Custom Q8 for Caption Comet for Asian and Middle Eastern languages
You don’t need to understand AI models. Choose the languages you need and Caption Comet tells you which optional model to download.
Caption Comet supports all transcription languages available through Whisper Large V3. Dual-model review is available for languages supported by a specialist model. Whisper can also translate supported spoken languages into English.
PRIVATE, LOCAL, AND BUILT FOR APPLE SILICON
Transcription happens locally. Your media does not need to leave your Mac.
Caption Comet supports Apple Silicon Macs with 8 GB of unified memory or more.
EXPORT
Export SRT, VTT, ASS, and TXT files, create MKV files with selectable subtitle tracks, or burn styled captions directly into finished videos.
FREE, TRIAL, OR LIFETIME PRO
Free Mode lets you transcribe and edit complete videos, with exports limited to the first 30 seconds and a Caption Comet watermark on burned video.
Every installation includes an optional 30-day Pro trial that begins only when you activate it.
Study Mode remains free without the 30-second transcript limit.
Upgrade to Lifetime Pro for unlimited exports, no Caption Comet watermark, and the complete Pro feature set.
One purchase. No subscription. No recurring fees.
Ratings & Reviews
- This app hasn’t received enough ratings or reviews to display an overview.
- New caption segmentation profiles: Choose Standard, Broadcast, or create custom profiles with sentence, word, character, line-count, and timing controls. Profiles work in both Quick Mode and Project Mode.
- Project Mode Reflow: Reformat an existing word-timed subtitle track without retranscribing. Reflow creates a separate track while preserving the original, including its timing and lineage.
- Broadcast output improvements: Added predictable 32-character line handling, one- or two-line captions, sentence-aware splitting, authored line breaks, and consistent SRT, VTT, ASS, MKV, burn-in, and TXT output.
- Better alternate-model review: Possible missing-passage or repeated-text warnings now show the actual passage in Project Review as well as Quick Mode. Each warning can be dismissed individually, or all warnings can be dismissed with Dismiss All.
- More reliable cancellation
- Safer timing and settings behavior
The developer, ALDO MANNINO, indicated that the app’s privacy practices may include handling of data as described below. For more information, see the developer’s privacy policy .
Data Not Collected
The developer does not collect any data from this app.
Accessibility
The developer has not yet indicated which accessibility features this app supports. Learn More
Information
- Seller
- ALDO MANNINO
- Size
- Up to 3.2 GB
- Category
- Photo & Video
- Compatibility
Requires macOS 26.0 or later and a Mac with Apple M1 chip or later.
- Mac
Requires macOS 26.0 or later and a Mac with Apple M1 chip or later.
- Mac
- Languages
- English
- Age Rating
4+
- 4+
- In-App Purchases
Yes
- 30-Day Trial for Mac $0.00
- Caption Comet Lifetime Pro $19.99
- Copyright
- © 2026 Aldo Mannino
