EmotionMap

EmotionMap

Abstract

This research explores the integration of affective computing into music production through the development of EmotionMap, an emotion-aware music production tool. The system analyzes vocal performances and automatically maps predicted emotional states to digital signal processing (DSP) parameters in real time. Using machine learning techniques, vocal acoustic features are extracted and used to estimate continuous emotional values within the Valence–Arousal (V-A) space. These emotional coordinates are transmitted to a custom Max for Live device in Ableton Live, where they control audio effects such as filter modulation and distortion. The system also supports parameter mapping to external plugins, enabling emotion-driven interaction across different production environments. Through the design and implementation of EmotionMap, this research investigates how affective computing and Music Information Retrieval techniques can support creative decision-making and enhance expressive possibilities in music production workflows.

Publication Date

7-1-2026

Campus

Valencia (Spain) Campus

Keywords

music emotion recognition; machine learning; Max for Live

EmotionMap

Share

COinS