Speak your video feedback. AI turns it into timeline markers.
Hold ⌃ + ⌥ (Ctrl + Alt) anywhere on the player to talk and sketch on the frame. AI transcribes your voice, detects spoken timecode ranges ('cut from 0:12 to 0:24') and pins searchable text to the exact frame.
At a glance
- Push-to-Talk voice notes
- Natural spoken timecode ranges
- Automated AI speech-to-text
- DaVinci & Premiere marker export
What you get
What you get with Voice Video Review.
Push-to-Talk voice notes
Hold ⌃ + ⌥ anywhere on the media player to speak notes and sketch simultaneously. Release to post immediately.
Natural spoken timecode ranges
Say 'cut from 0:12 to 0:24' to automatically bind comments to timeline ranges with seekable interactive chips.
Automated AI speech-to-text
Every voice note is transcribed into searchable text so editors get a clear, readable checklist.
DaVinci & Premiere marker export
Export all transcribed voice notes straight to DaVinci Resolve (EDL) and Premiere Pro (XML) without plugins.
Push-to-Talk (⌃ + ⌥) Instant Voice Notes
Hold two keys anywhere on the media surface, speak feedback, sketch with arrows or freehand brush, and release. No separate upload or recording setup.
Natural Spoken Range Extraction
Say 'cut from 0:12 to 0:24' or 'remove from 1:04 to 1:18'. Dusken's speech engine recognizes timestamps and automatically creates interactive range chips.
Searchable Transcripts & NLE Export
Every voice note is transcribed into searchable text. Export notes straight to DaVinci Resolve (EDL) and Premiere Pro (XML) as readable timeline markers.
How it works
How voice-powered video review works.
From natural speech to frame-accurate timeline markers in your NLE.
Hold Push-to-Talk (⌃ + ⌥)
Press and hold ⌃ + ⌥ anywhere on the media player to activate the microphone and sketch tools simultaneously.
Speak and sketch on the frame
Point out pacing issues, color shifts, or sound design notes while drawing directly over the video frame.
AI transcribes and pins the comment
Release the keys. AI automatically transcribes your audio into searchable text and attaches the visual drawing to the exact timestamp.
Export to your editing timeline
Export all transcribed voice notes to DaVinci Resolve, Premiere Pro, or Final Cut Pro as frame-accurate sequence markers.
The shift
What changes with Dusken.
Typing three paragraphs into a text box trying to describe visual pacing, or recording a 15-minute Loom video that your editor has to scrub through.
Hold ⌃+⌥ to speak and sketch simultaneously. AI transcribes your audio into frame-pinned text markers with seekable timecodes.
| Capability | Traditional Process | With Dusken |
|---|---|---|
| Review & Approval Flow | Typing three paragraphs into a text box trying to describe visual pacing, or recording a 15-minute Loom video that your editor has to scrub through. | Hold ⌃+⌥ to speak and sketch simultaneously. AI transcribes your audio into frame-pinned text markers with seekable timecodes. |
| Stakeholder Access | Expensive per-seat licensing or account friction required to view drafts | Zero-login instant browser links, 100% free unlimited reviewers on every plan |
| Precision & Tools | Vague timestamp typing, miscommunicated edits and lost version history | Frame-pinned annotations, voice-note transcription, multi-mode version compare and NLE marker export |
“Describing a motion design cut in a text box took 20 minutes and still got misread. With Push-to-Talk, I just speak and draw, and my editor gets exact timecode markers in seconds.”
Creative Director
Why creative teams switch to voice video review.
10x faster than typing
Speaking takes seconds compared to typing long paragraphs trying to describe visual creative nuances.
Zero ambiguity for editors
Combining tone of voice, visual arrows, and frame-accurate timestamps eliminates revision confusion.
Searchable text timeline
Editors don't have to listen through long audio files; AI transcription provides a readable, filterable checklist.
Questions before you start.
Hold ⌃ + ⌥ (or Ctrl + Alt on Windows) anywhere on the video player to speak and sketch on the frame simultaneously. When you release, Dusken transcribes the audio into searchable text and pins it to the exact frame.
No. Dusken transcribes all voice and video feedback automatically into text markers. Editors can read the comments as text, search by keyword, or listen to the original audio recording if desired.
When you speak time references like 'cut from 0:12 to 0:24' or 'fix lighting between 1:05 and 1:15', Dusken's speech engine automatically detects the range and binds the comment to that timeline segment.
Yes. Clients open the review link in any browser, click to speak or type, and their voice feedback is transcribed with zero login or software installation required.
Yes. You can export transcribed voice comments as CMX3600 EDL for DaVinci Resolve, XML for Premiere Pro, and FCPXML for Final Cut Pro.
Got something else? Contact us
We actually read every message. Bugs, ideas, rants, all welcome.
Have questions or want to see how Dusken fits your workflow?
Dusken is founder-led, so support is close to the product. We are happy to talk through your agency setup, NLE export workflows or custom requirements.