Quickstart guide

Get going fast. Get great results.

The habits that make Taperat noticeably smarter, and a checklist for your first 15 minutes.

Best practices

01

Start small.

Before importing your entire collection of recordings, try a game or two. Transcribe them, or run the Perception Engine on them, to get to know the layout of the app. If you have used a video editor like DaVinci Resolve or Premiere Pro, you will feel right at home.

Most important habit

Perceive. Correct. Save hours for future you.

After every Perception Engine run you get a summary, where you can correct character detections and scene findings, and look through the transcript and the diarisation (who spoke when). Don’t skip it, and where you can, fix the outliers.

When it asks “Is Speaker 2 Elara?”, a quick Keep or Not this teaches it more than any setting. It will save you hours of work later.

02

The library is only as smart as you let it be.

Voice fingerprints get better the more you teach them. Start with scenes where fewer than four characters talk, and correct what the Perception Engine detected there. Every correction helps it recognise your characters later.

Once a voice has a fingerprint, Taperat picks it out of bigger groups too, and everything else improves with it.

30 sof hand-corrected speech already helps massively
60 smakes detection much more reliable
03

How should you play?

The Perception Engine already recognises characters when others call them by name. To make detection golden, play with subtitles on, ideally with speaker names shown, if the game offers it.

Don’t want to play the whole game like that? A handful of clips with subtitles is already enough to pin a voice to a name.

04

Let the dialogue be heard.

If the game lets you, turn music and effects down a little and keep voices up. Clear speech means cleaner transcripts and stronger fingerprints.

05

Agentic workflowStudio

Using an external agent locally? Great! You can use any agent you like through MCP Bridge. Clearly named and tagged slices and recordings let your agent find exactly what it needs and place it straight onto a DaVinci Resolve timeline.

So give slices clear titles and put them in collections as you go. Good names help your own searches just as much.

Connect MCP Bridge

  1. 1Start Taperat. The bridge works through the running app, so keep it open; the tray is fine.
  2. 2Perceive first. Run the Perception Engine on a recording before handing it to the assistant: the transcript and the separated voices are what it works from. MCP Bridge builds on your library; it doesn’t replace it.
  3. 3Open Settings › Assistant › Connect a client and copy the snippet: a block for Claude Desktop or one line for Claude Code. Any MCP client works the same way.
  4. 4Tell it how deep to go: quick, thorough or exhaustive. That sets how many frames it may look at, and it tells you what it used.
  5. 5Ask in plain words, then review. Everything it makes is marked as the assistant’s, and one undo takes the whole session back.

What makes a good ask

  • Name the recording, session or bin.
  • List the kinds of moments you want, then any one-off moments separately.
  • Say who is in them.
  • Name the collections and tags you want. It reuses the ones you already have.
“In this session from game X, make slices of every battle and every story beat with character Y. Separately, find the scene where character Z betrays character Y. Put them in the Story and Combat collections and tag the characters.”

Connecting an external agent or AI assistant that runs in the cloud shares your files with that service. For footage under NDA, keep MCP Bridge off or use an assistant that runs on your PC.

Your first 15 minutes

  1. 1Add one recording folder.Full guide
  2. 2Transcribe one recording.Full guide
  3. 3Search for a line you remember.Full guide
  4. 4Save your first slice: I, O, Enter.Full guide
  5. 5Export it, or send it to your editor.Full guide

Want the details?

Every feature, step by step.

Open the step-by-step guides