01 Transcription and dictation
Accurate, editable subtitle extraction
Choose Parakeet, Qwen3 ASR, FireRed, or Cohere models for local audio/video transcription. Add custom vocabulary for specialist terms.
English
Built for Mac audio and video creators
Local Mac transcription, captions and AI rough cuts.
Cliptate transcribes audio and video, trims pauses, and edits captions locally on your Mac. It also supports dictation and text to speech. After model downloads, local speech works offline without media uploads.
Apple Silicon · macOS 15+ · Free GitHub build
Core capabilities
One on-device AI toolkit built for creator workflows.
Import media, edit segments through their transcript, then export captions, audio, video, or a timeline. Manage models, vocabulary, and configuration presets in the app.
02 Smart trimming
Trim pauses in bulk and shape rough cuts faster
Auto-detect silence and bulk-trim dead air from podcasts, interviews, and recordings.
03 Speaker diarization
Auto-identify speakers and label dialogue
Use Sortformer to distinguish speakers in supported transcription pipelines. Silero VAD detects speech activity, making interviews and meeting recordings easier to organize.
04 Voice synthesis
From narration to original voices
Qwen3 TTS offers preset voices, reference-audio cloning, emotion control, and voice design. Available controls depend on the selected model.
05 Correction and translation
Correct transcripts and create bilingual captions
Configure an AI provider and API key for correction and keyword emphasis, including custom OpenAI-compatible endpoints. Apple Translation handles captions for supported language pairs.
06 Visual and caption effects
Give spoken content rhythm and emphasis
Configure filler-word filtering, scaling, and zoom. Extract highlights with your configured AI provider, and set landscape/portrait caption styles and translation placement.
Voice capabilities
Live dictation and local voice synthesis.
From live dictation to TTS dubbing — two voice capabilities, entirely on-device.
Voice dictation
Speak to type — editable text in real time
Turn microphone audio into editable text, with custom vocabulary and dictation history. GitHub builds support global shortcuts and focused-field insertion after Accessibility permission is granted.
- Local recognition; streaming depends on the selected model
- Review, copy, and manage dictation history
- Custom vocabulary for domain-specific accuracy
- Without Accessibility permission, copy and paste text
TTS voice synthesis
Choose voices and styles by model
Preset
Preset voices
Download the matching model, then select a preset speaker for narration.
Clone
Voice clone
Import local reference audio and its transcript to clone a voice with a Base model.
Emotion
Emotion control
Add emotion and style prompts to named speakers with a CustomVoice model.
Design
Voice design
Describe a voice in natural language and create it with a VoiceDesign model.
Workflow
From import to export, keep the workflow in one fast production line.
The most time-consuming creator tasks, compressed into four connected steps.
01
Import common audio and video formats
Drop podcasts, vlogs, interviews, or meeting recordings straight into the workspace.
02
Transcribe, then edit through text
Trim pauses with silence detection, review speakers and segments, edit text, and optionally run AI correction.
03
Create multilingual dubbed and translated versions
Translate captions, then generate dubbed audio with TTS. Use an automatically extracted source-audio reference or your own reference recording.
04
Export captions, media, and timelines
Export FCPXML, SRT, ASS, TXT, video, or audio. Combine the working set or export per video, with optional burned-in captions.
Privacy and performance
Local speech processing. Online features are your choice.
Transcription, silence detection, diarization, and TTS run on your Mac. Online AI features send relevant text and any context you choose to the configured service.
- Local speech inference needs no media upload or AI API key.
- Model downloads need internet access; local speech features work offline once cached.
- AI correction, keyword emphasis, and dictation enhancement use your configured provider; online services may charge separately.
- For sensitive content, use local speech features and Apple Translation; language packs may need downloading first.
Professional output
Deliver the right formats for post-production.
AI Cut exports timelines, captions, audio, and video. The CLI also provides JSON, Markdown, and VTT for automation.
FCPXML Timeline project exchange
SRT Universal subtitle delivery
ASS Styled subtitle support
TXT Timestamped transcript
FCPXML targets Final Cut Pro with a Resolve compatibility option. Import captions in formats your editor supports. Video: MP4 / MOV / M4V. Audio: M4A / WAV.
Use cases
Ready for podcasts, vlogs, courses, interviews, and multilingual content.
Podcasts and interviews
Trim dead air, add captions, and identify publishable moments faster.
Vlogs and short-form video
Make talking-head edits smoother and keep attention on pacing and expression.
Courses, meetings, and dictation
Turn long recordings or live voice input into searchable, translatable, and deliverable text assets.
Multilingual dubbing
Ship alternate language versions faster while preserving expressive tone.
Workflow guides
Articles & comparisons
From dictation to subtitle delivery: compare Cliptate with other tools by processing, cost, and everyday tasks.
Download and install
Choose your Mac build.
Requires Apple Silicon (M1 or newer) and macOS 15 or later. Model downloads and memory requirements vary; allow enough storage for your selections.
Free stable release
All models and features are free, with no subscription or purchase. Includes the native CLI and in-app updates, plus Accessibility-backed text insertion and global shortcuts with permission. Online AI providers charge separately.
Open GitHub ReleasesNightly preview builds
Nightly builds track main periodically and may contain unreleased changes. Check each build’s notes for included features.
Nightly preview builds