[

MapifyAI

From a Audio or Transcription to Mind-Map Diagram with this AI-powered tool

This project has been submitted to AssemblyAI Challenge : Sophisticated Speech-to-Text.

What I Built

An Application that allow from an Audio concerning a discussion, a meeting, etc ... to generate a "meaningful mind-map diagram", that represent the touched key points. This representation joined with summary provide a more complete and understandable informations

Demo

The application in available here for access to full functionality you need both an AssemblyAI Api Key and a OpenAI Api Key. Below there are some representative screenshots

Settings

Upload Audio

Transcribe Audio

Generate Mindmap Diagram

Journey

To implement process from audio to diagram I have create several skilled agents described below:

1. transcribe-from-audio: this agent use AssemblyAI transcripts API to transcribe the provided audio.

2. keypoints-from-transcript: this Agent use OpenAI (got-4o-mini) to extract the Keypoints inside the given transcription

3. summary-to-mindmap: this agent use OpenAI (got-4o-mini) to arrange the key points in a kind of ontology providing a hierarchical representation of information

4. mindmap-to-mermaid: last agent transform the mind-map representation in a mermaid syntax ready for the visualization

Diagram of Agentic Architecture

graph TD;
        __start__([<p>__start__</p>]):::first
        __end__([<p>__end__</p>]):::last
        transcribe-from-audio(transcribe-from-audio)
        keypoints-from-transcript(keypoints-from-transcript)
        %%keypoints-from-audio-transcript(keypoints-from-audio-transcript)
        summary-to-mindmap(summary-to-mindmap)
        mindmap-to-mermaid(mindmap-to-mermaid)
        keypoints-from-transcript -->|Keypoints summary | summary-to-mindmap;
        mindmap-to-mermaid -->|Mermaid script| __end__;
        summary-to-mindmap -->|Mindmap structure| mindmap-to-mermaid;
        transcribe-from-audio -->|Transcription| keypoints-from-transcript;
        __start__ -.->|Audio| transcribe-from-audio;
        __start__ -.->|Transcription| keypoints-from-transcript;
        %%__start__ -.-> keypoints-from-audio-transcript;
        %%__start__ -.-> summary-to-mindmap;
        %%__start__ -.-> mindmap-to-mermaid;
        %%__start__ -.-> __end__;
        classDef default fill:purple,line-height:1.2;
        classDef first fill-opacity:0;
        classDef last fill-opacity:0;

Loading