New Dimensions in Sound

By Sam Shin

Audiovisual

New Dimensions in Sound explores the intersection of artificial intelligence and human creativity in both music and visuals. The piece uses Google AI’s Magenta Studio, a suite of tools for Ableton Live, to generate melodies and drum patterns, forming the musical backbone of the work.

Visually, the piece draws from a 1957 RCA educational video titled How to Listen To... New Dimensions in Sound, which introduced the public to high-fidelity audio technology. By reworking this archival footage through an AI text-to-video model called TempoFunk, the original material is transformed into new, abstract sequences. Each section of the video was generated using written prompts and still images from the RCA footage, with dialogue from the original video serving as creative inspiration.

While more advanced generative models have emerged since this work was created, New Dimensions in Sound presents a unique method of artist-AI collaboration. Rather than allowing the AI to generate complete works autonomously, the artist actively selects, refines, and arranges AI-generated content, much like a composer or editor. This process emphasizes artistic intention, demonstrating how AI can be a creative partner rather than just a tool.