A little less work. A little more you.
How local video transcription and SRT exports work
Understand browser transcription, SRT versus burned-in captions, model downloads, accuracy limits, and optional transcript sharing in Kitecut.
By Kitecut · Published · Updated
Quick answer
Kitecut downloads a speech model and transcribes the audio in your browser. SRT and VTT are separate subtitle files; burned-in captions become part of the video picture. Review automatic text and timing, and allow for the memory and processing limits of your device.
The model downloads; the original media stays on your device
Kitecut downloads a supporting speech model, then processes speech locally in the browser. The download service receives the network information needed to deliver those files.
This still needs device resources. Long recordings, limited memory, and browser restrictions can affect completion time. Start with a short spoken clip when trying a new device.
Choose SRT, VTT, or burned-in captions
SRT and VTT are separate text files containing subtitle cues and timing. They can be loaded by compatible players or publishing tools. Burned-in captions become part of the picture and cannot be switched off like a subtitle track.
The video-to-SRT starter creates separate SRT subtitles with burning disabled. Review the captions step before running if you need a different output.
Review every automatic transcript
Speech recognition can confuse names, technical words, accents, or simultaneous speakers. Quiet speech and music can also affect the result. A generated transcript is a draft to review.
Listen alongside the subtitles and correct text and cue timing in your subtitle workflow before publication. Keep the original recording so mistakes can be checked later.
Sharing a transcript is a separate choice
Local transcription does not automatically send the words to hosted AI. Enabling transcript sharing lets the assistant use those words, for example to help plan a content-based cut.
AI messages, technical media details, and an explicitly shared transcript are processed by external services. Use the editor's privacy receipt to inspect what was sent. Local tools remain usable without sending a chat request.
Put it into practice
Generate an SRT subtitle file locally
Transcribe a video's speech on your device and export SRT subtitles. No original media upload; review automatic transcription for mistakes.
Extract MP3 audio from a video
Save a video's audio track as MP3 in your browser. Process a local file on your device and download the audio without a watermark.
Trim a video online, on your device
Keep the part of a video you need. Adjust start and end times in your browser, preview your plan, and download a watermark-free clip.