AI transcription with speaker labels, text-based editing, animated captions, Smart Clean, remote recording — all offline, one-time price. No cloud, no subscription, no data leaving your machine.
Text-based editing
Delete words to cut, mute, adjust gain, or change speed — all by editing text. Full keyboard shortcuts with custom remapping. 100-level undo/redo.
Handles 3-Hour Podcasts
No lag on 3-hour podcasts — virtualized transcript and chunked waveform. Speaker diarization labels who said what. Pause and resume transcription.
Export how you want
MP4, MOV, WebM, WAV — up to 4K. Background removal, video zoom, animated captions, chapters, background music with auto-ducking, crossfade at cuts, and plain-text transcript export. Fast stream-copy for unchanged segments.
Script Mode
Write a script, record yourself reading it, and get line-by-line alignment. Perfect for voiceovers, narration, and matching recorded audio to a written transcript.
AI transcription & editing
Transcribe video to text with speaker labels using Whisper AI. Remove filler words (um, uh, like), Smart Clean for one-click audio cleanup, AI rephrasing, social clip suggestions, and show notes generation. Custom glossary for domain-specific terms. Export transcripts as SRT, VTT, ASS, or plain text.
AI Chat Assistant
Chat with your transcript. Get rephrases, summaries, show notes. The assistant executes edits directly. Supports Ollama, OpenAI, Claude.
Smart Clean & Filler Removal
One-click audio cleanup: automatic filler word removal (um, uh, like, you know) + silence trimming + noise reduction + loudness normalization. Presets for Podcast, Interview, Social, and Presentation. No cloud processing required.
Animated Captions & SRT Export
Burn in TikTok-style word-by-word captions with presets (TikTok, YouTube, Clean) or full manual customization. Export captions as SRT, VTT, ASS, or plain text — perfect for YouTube SEO, social media clips, and accessibility. No cloud processing needed.
Text-to-Speech
Multi-engine TTS: OmniVoice (600+ languages, voice cloning) and Kokoro-82M (54 preset voices, CPU-friendly). Generate voiceovers entirely offline. No cloud API needed.
Privacy first
100% offline. No account, no cloud, no telemetry. Auto-save crash recovery with resume on relaunch. Your data stays on your machine.
Remote Recording
Record remote podcast interviews in high quality. Host records locally, guest joins via a shareable link. Both tracks merge into stereo WAV (host L, guest R). No cloud dependency.
