EditNova
Video Editor
Pick a video, say what you want - or add steps yourself. Everything runs in this browser.
Intro / outro, extra tracks & special effects (all optional)
Intro / outro clips (your own files)
Pick a video or image you have permission to use. Intro goes first, outro goes last. Images start at 3 seconds; edit their duration below. These are real timeline clips, not stock templates. Undo removes the last import.
Independent visual and audio tracks
Initial renderer: up to 4 extra visual clips and 4 extra audio clips. Short desktop tests only; phone and long-edit capacity is not verified. Add files, then set source trim, final-video timing, position, size, mute/solo in the Plan. These do not ripple when base clips change. The source preview does not show these effects; render to check. Files are saved only on this device, per account.
Drag a bar to move it; drag its right edge to change duration. Numeric timing in the Plan is also available. Later visual Plan steps appear on top. Solo affects sound only.
Original sound effects
Three simple effects synthesized on this device, not sampled stock recordings. Preview before adding. Each uses one of the 4 extra audio clip slots; set its timing and gain in the Plan. These are not realistic Foley or a stock sound library.
Word captions (automatic transcription)
Transcribe ONE chosen video on this device, then check and correct every word and timing before applying. Long videos are processed section by section in real time, so they take about as long as the video itself plus recognition time; keep this tab open and the screen on. One honest bound: the review screen holds up to 6000 words in this version (roughly an hour of speech) - past that, transcription stops and keeps a usable partial draft of the first part. Nothing is uploaded. You can also import reviewed word data (JSON) for one unchanged source video.
First transcription downloads a large model (about 135 MB for Roman, about 200 MB for Hindi script) - use Wi-Fi; the browser may need to re-download it later. Large model; may not run on low-memory phones. Nothing loads until you tap transcribe.
Text behind a person / in front
Experimental local person-mask render, not lip-reading or automatic captions. One close-up person works best; hair, hands, movement and multiple people can have errors. Short draft: up to 4 seconds, 6 fps, at most 360 pixels on the long side. Original clip is untouched. Review before adding. No source upload.
This draft is flattened video, not editable text layers. Keep your source. Audio is re-encoded from the selected source interval when available. Source frame seeking is approximate. This is not full-quality export or a phone-capacity guarantee.
Person model and licenseDrop files on the timeline to add clips.
All editing steps (27 tools - everything the editor can do)
Freeze frame (pause on one frame)
Splits the picked video at that second and holds the frame as a still photo for the chosen time. Works on the first picked video. Sound pauses during the freeze. Use Undo to reverse it.
Review, music, voice-over, logo & other extras (all optional)
Composited picture and mix review
Render-backed review, not a live effects preview. Uses the same full render as export, including extra tracks, stacking, timing and mute/solo. Rendering takes time and memory. It does not replace the source preview above.
Add music? (optional)
Record a voice-over? (optional)
Add a logo/watermark? (optional)
Subtitle file? (optional - .srt file; auto captions: see Word captions in the previous group)
Color LUT file? (optional - .cube)
Green screen background? (optional - image)
Background photo - person tool (optional)
Used by the Background tool when its mode is "My photo": the person in your video is placed over this photo. Nothing is uploaded.
Add a video/image overlay? (B-roll, optional)
For stickers, pick your own transparent PNG/WebP here. This is one shared overlay slot, not a multi-sticker library.
Replaces the picked overlay after confirmation. Shape uses the existing Overlay step's size, corner, timing and opacity. Download generated shapes to keep their files; project snapshots only keep the file name and settings.
Saved projects, backup & history
Backups contain edit settings and file names, not source media. Opening a backup replaces the current edit after confirmation; Save creates a new project.
Saved version history
Up to 5 saved snapshots, 250 KB combined; older or oversized entries are not retained. Edit settings and file names only, not media. Load for review does not save automatically.
- This is real editing: trim, join, reel sizes (9:16/1:1/16:9), blur background, photo clips (slideshow), compress, mute, speed, multiple texts at different times, logo, color & brightness, volume/fade, music, MP3 audio, thumbnail - all running in your browser. Videos are never uploaded.
- Long videos take time (a 1-minute video can take a few minutes on a phone). Files over 300 MB may struggle.
- Not yet: auto best-scene selection, AI voiceover, direct upload to Instagram/YouTube (we will ask separate permission before connecting accounts). These will appear only when they truly work.