Inspiration
We are interested in providing tools for humans to improve their habits by providing insights into their behavior using various biometrics.
What it does
This specific application uses the Muse EEG headband and the Hume emotional facial expression analysis to watch users as they play chess.
How we built it
We have created an MCP server that records EEG brainwave data and also processes video input using the Hume facial emotional analysis model. The MCP server is then plugged into our voice bot that uses and llm to interface with users via voice and allows them to interact with the readings and analyze the data.
Challenges we ran into
Handling large amount of real time data both from video and EEG samples .
Accomplishments that we're proud of
We were able to get the video and the eeg data end-to-end from Muse device and camera all the way to LLM and then have the data ready to be analyzed by the LLM.
What we learned
The MCP interface needs to be very descriptive and guide the LLM on the data.
What's next for drishtisutra
We have plans to take it to a consumer grade mobile app that can be used in various environments to help users improve their life.
Built With
- amazon-web-services
- cartesia
- fishaudio
- node.js
- pipecat
- python
- sql
- typescript
Log in or sign up for Devpost to join the conversation.