---
title: "Search Audio and Video · AssistantGeneral"
description: "Recordings and videos are transcribed on your machine and searched in 30-second moments. Answers cite the minute and second, and a click plays it from there."
url: "https://royalsoftworks.com/products/assistant-general/features/audio-and-video/"
source: "https://royalsoftworks.com"
format: "markdown"
note: "Markdown rendering of the HTML page at `url`. Same content, same canonical URL."
---

Features · Your documents

# Find the moment in a recording.

Recordings and videos are transcribed on your machine and searched in 30-second moments. Answers cite the minute and second, and a click plays it from there.

[Editions](https://royalsoftworks.com/products/assistant-general/editions/): Every edition. Seeing what is on screen needs an optional download.

## Recordings and video, feature by feature

Each comes with a small drawing of how it works. They are diagrams, not screenshots; numbers in them are examples.

- ### Moments, not whole files

  A recording is cut into 30-second moments by default, each searched on its own. The length is adjustable.

  Timeline (30-second moments): 0:00, 0:30, 1:00, 1:30, 2:00.
- ### Transcribed on your computer

  Speech becomes text line by line, each line tagged with its time. No audio leaves the machine.

  For your IT team: Whisper, run on the processor.

  Diagram: Audio → Whisper (processor) → \[0:42\] a line.
- ### What is on screen

  With the optional download, frames are described in words by the vision model, so a moment can be found by what is shown.

  Diagram: A frame every 2 s → Vision model → “A chart of Q3 sales”.
- ### How it sounds

  The default search model also reads sound, so a moment can be found by what it sounds like.

  Diagram: Sound → Search model → Found by sound.
- ### Your choice of detail

  How often a frame is looked at, how much picture detail is kept, and whether the graphics card helps.

  Diagram: Frame interval, Low / Normal / High, Graphics card.
- ### Cited to the second

  Answers cite the minute and second. A click plays the recording from that moment in the app's own player, or opens it in yours.

  Diagram: Answer \[1:42\] → Click → Plays from 1:42.
- ### YouTube by the moment

  Video transcripts are kept in timed windows too.

  Timeline (Transcript in timed windows): 0:00, 0:30, 1:00, 1:30.

## Goes well with

- [Files it reads](https://royalsoftworks.com/products/assistant-general/features/file-types/)
- [Search models](https://royalsoftworks.com/products/assistant-general/features/embedding-models/)
- [Voice](https://royalsoftworks.com/products/assistant-general/features/voice/)

[Previous Files it reads](https://royalsoftworks.com/products/assistant-general/features/file-types/) [Next Knowledge map](https://royalsoftworks.com/products/assistant-general/features/knowledge-insights/)

[All 31 feature classes](https://royalsoftworks.com/products/assistant-general/features/)

<!-- verse:royalsoftworks
     Psalmi 55:17
     Večerom i jutrom i u podne tužim i uzdišem, i čuće glas moj;
     https://www.wordproject.org/bibles/verses/serbian/02_prayer.htm
-->
