VBox
VBox is an AI-powered radio for musical exploration and group vibrations.
Team Project
2023 Spring, 5 weeks
Interaction Intelligence (MIT)
Info:
Team:
Artem Laptiev (MIT)
Skills:
Industrial Design; Electronics; Prototyping
Tools:
Solidworks; Illustrator; Laser Cutter; 3D Printing; Arduino
I worked on the hardware/electronics design. My responsibilities include sketching, CAD modeling, fabrication, and visual design. As a team, we collaborated on the brainstorming and integration of the prototype.
Role:
Context
How do we utilize the large language model to capture the abstractions of a musical experience and take the audience on an interactive musical journey through empathy?
Integration of AI into sound objects
In a world of endless musical complexity, using generic terms to describe the sound is a crime. Is the track just happy? Is it filthy? Purple? Carpe diem? *^5#@**? Through the Large Language Model, VBox captures the various abstracted understandings of the song played in the form of texts and lets you travel down one of these rabbit holes to find more of what is hard to describe.
Inspiration
AI and Musical Creation
We wanted to explore how AI’s capability in music creation and interpretation may be translated into physical interfaces. Particularly the intersection between AI and nostalgic music interfaces, such as cassette tapes and record players that store and extract music from analog cartridges.
Brainstorming
I brainstormed for the features and UI configurations, including a genre selector and adjustable parameters for the song’s popularity, tempo, and energy.
We decided to focus on integrating natural language into the musical experience and create a more minimal user flow paired with a form that resembles a traditional radio. It includes a display, two simple buttons, a crank that interfaces the display, and a cartridge system.
System Diagram
Prompt Engineering
Prototyping
After prototyping the different visions for VBOX, We decided to simplify the interface by prioritizing one interaction point that is more playful and experimental over hard-coded buttons.
Experimenting with sensors and trying ways to automate the crank and explore any potential opportunity for decision-making exchange between AI and user.
