Turn audio recordings into mind maps

SpawnGraph does not transcribe audio — let us be upfront about that. What it does is take a transcript and structure it into a visual mind map that you can actually navigate. The workflow is two steps: transcribe your audio with any tool you already use, then paste the text into SpawnGraph. The whole thing takes a few minutes and turns an hour of audio into a map you can share in a Slack message.

The workflow

  1. 1Transcribe the audio. Use Whisper (free, runs locally), Otter.ai, your meeting platform's built-in transcription (Google Meet, Zoom, Teams), or any other transcription tool. Export or copy the text transcript.
  2. 2Paste the transcript into SpawnGraph. SpawnGraph's in-browser NLP detects topic shifts, speaker turns, and structural markers in the text and organises them into a hierarchical map.
  3. 3Edit and share. Rename branches, highlight action items, collapse the tangents, and share the map link with everyone who was in the meeting — or everyone who was not.

Which transcription tool should you use?

It depends on your setup. Whisper is the best free option — it is open-source, runs on your own machine, and produces accurate transcripts for most audio quality levels. Good for privacy-sensitive recordings. Otter.ai works well for live meetings and produces speaker-labelled output that maps cleanly. Google Meet and Zoom both have built-in transcription in their free and paid tiers — if you already recorded the meeting there, just download the transcript. For podcasts and lecture recordings, Descript produces clean structured output. The point is: use whatever you already have. SpawnGraph takes the text from any of them.

Why a map beats a transcript wall

Transcripts are hard to navigate. An hour-long meeting produces eight to twelve thousand words of unstructured text. You cannot skim it the way you can skim a document. A mind map gives you the navigation layer that transcripts lack: topics at the top level, sub-points underneath, decisions and action items pulled out into their own branches. You can share a link to the map and someone can understand the full meeting in three minutes instead of reading for forty.

Lecture recordings and podcast interviews

This workflow is not just for meetings. Lecture recordings are common for students who want to process audio from a class they attended or watch again later. Podcast interviews — especially long-form conversations — often contain dense information that deserves a structural overview. Research interviews that you have recorded and transcribed for qualitative analysis map well too: themes emerge naturally when you run the transcript through SpawnGraph.

Transcript text stays in your browser

When you paste a transcript into SpawnGraph, the text is processed in-browser using client-side NLP. It is not sent to a server and not retained anywhere. Meeting transcripts often contain confidential discussions; the privacy model here is as strong as the local parsing — nothing leaves your tab.

In short: Transcribe your audio with Whisper, Otter, or your meeting platform. Paste the transcript into SpawnGraph. Get a navigable, shareable mind map in seconds. Free, no signup, nothing uploaded.

Frequently asked questions

Can SpawnGraph convert audio directly to a mind map?
SpawnGraph does not transcribe audio itself — it processes text. The workflow is: transcribe your audio with a tool like Whisper, Otter.ai, or Google Meet, then paste the transcript into SpawnGraph. SpawnGraph handles the visual structuring step.
What transcription tools work well before SpawnGraph?
Whisper (free, open-source, runs locally) is excellent for privacy. Otter.ai is good for meeting recordings with speaker labels. Google Meet and Zoom both have built-in transcription. Any tool that outputs a text transcript will work with SpawnGraph.
How does SpawnGraph handle meeting transcripts?
Meeting transcripts typically have speaker labels, timestamps, and conversational back-and-forth. SpawnGraph detects topic shifts and groups related exchanges into branches. The result is a topic-organised map of the meeting rather than a flat wall of dialogue.
Does the transcript text get uploaded to SpawnGraph servers?
No. Pasted text and dropped files are processed entirely in your browser using client-side NLP. The content never leaves your device.
What types of audio recordings work well for this workflow?
Meeting recordings, lecture recordings, podcast interviews, research interviews, and conference talks all work well. Transcripts with clear topic structure produce the cleanest maps.

Related tools

Start mapping in seconds — free

No account required for basic use. No data leaves your browser.