Being Good was a long-running effort to develop a different way of describing and navigating music. Rather than relying primarily on genre, artist similarity, or metadata, we explored how Rhythm, Texture, and Pitch could describe perceptual relationships between music and mood. I worked on the project from 2013–2024 as a co-inventor, across multiple generations of research, technology, and interface design.
My work existed between music research and product experience: perceptual evaluation, music analysis, interaction and UX design, iterative prototype development, and translating the underlying system into something people could actually use / demonstrate. I evaluated music, compared model output, tested prototypes and product flows, and helped shape the system through more than 1,100 personal song evaluations, multiple working prototypes, and the research that contributed to its issued U.S. patents.
Music information retrieval · RTP classification · Acoustic feature analysis · Machine learning
One of the hardest parts was turning subjective listening into data that could actually be tested. During the Multimood phase, I manually evaluated 1,123 musical segments using half-point Rhythm, Texture, and Pitch values; roughly 30% became training data, while the remaining 771 segments were used to compare computed RTP and mood classifications against my perception. Later versions brought that research into working software for analyzing music, classifying and filtering by mood and intensity, generating playlists, and connecting results back to Spotify-based playlists as a demo source.
The work ultimately produced a continuing patent family, including US 9,875,304 B2, Music Selection and Organization Using Audio Fingerprints (2018), and US 11,899,713 B2, Music Streaming, Playlist Creation and Streaming Architecture (2024), on which I am named as an inventor. Rather than ending as one definitive consumer product, Being Good produced a body of music-perception research, a manually labeled dataset, multiple working software systems, several product directions, and patented technology.
It also changed how I think about design: sometimes drawing the interface is the easy part. The harder problem is deciding what the system means in the first place.
