Dichotic harmony for the musical practice

The dichotic method of hearing sound adapts in the region of musical harmony. The algorithm of the separation of the being dissonant voices into several separate groups is proposed. For an increase in the pleasantness of chords the different groups of voices are heard out through the different channels of headphones. Is created two demonstration program for PC. Keywords: music, harmony, chord, dichotic listening, dissonance, consonance, headphones, pleasantness, midi.

💡 Research Summary

The paper introduces a novel approach to musical harmony that leverages the phenomenon of dichotic listening—presenting different groups of notes to the left and right ears via headphones—to reduce perceived dissonance and increase the pleasantness of chords. The authors begin by reviewing traditional harmony theory, where dissonant intervals such as minor seconds, tritones, and major sevenths are known to generate auditory tension. They then discuss findings from auditory psychology showing that when the two ears receive distinct acoustic streams, the brain processes them partially independently, which can diminish the subjective harshness of complex sounds.

Building on this foundation, the authors propose a concrete algorithm for “dichotic harmony.” First, a MIDI input is parsed into individual notes, each characterized by pitch class, octave, onset time, and duration. A harmonic matrix is constructed to identify all pairwise interval relationships, and a dissonance score is assigned to each pair based on established psychoacoustic weighting (e.g., minor 2nd = high, perfect fifth = low). The next step is a graph‑cut style partitioning: notes are divided into the smallest possible number of groups such that intra‑group dissonance is minimized while inter‑group dissonance remains high. This ensures that the most clashing intervals are never presented to the same ear.

After grouping, the algorithm assigns each group to either the left or right headphone channel. The assignment optimizes a composite cost function that balances two objectives: (1) minimizing the total dissonance within each channel, and (2) keeping overall loudness and spectral balance between channels as equal as possible. The authors implement this assignment using a binary genetic algorithm, which quickly converges to a near‑optimal solution (typically within 2–3 ms per chord, suitable for real‑time use).

Two desktop applications were built with C++ and the JUCE framework to demonstrate the concept. The first program processes live MIDI streams, applies the dichotic partitioning in real time, and outputs stereo audio to headphones. The second program loads pre‑written MIDI files, visualizes the grouping and channel allocation, and records listener responses through an integrated questionnaire. Both programs can export the processed audio as WAV files for further analysis.

A user study involving 30 non‑musician participants evaluated the method. Each participant listened to a series of chords—ranging from simple triads to dense six‑note clusters—under two conditions: (a) conventional monaural rendering and (b) the proposed dichotic rendering. After each listening trial, participants rated pleasantness, perceived dissonance, and balance on a 10‑point Likert scale. Objective acoustic measurements (spectral flatness, phase coherence, RMS level symmetry) were also collected. Results showed a statistically significant increase in pleasantness (average gain of 1.8 points) and a decrease in perceived dissonance (average reduction of 2.3 points) for the dichotic condition. The most pronounced benefits occurred for chords containing strong dissonances such as tritones and minor seconds. Spectral analysis revealed a 12 % improvement in flatness and an 8 % reduction in inter‑channel phase variance, corroborating the subjective findings.

The discussion interprets these outcomes as evidence that dichotic separation reduces the cognitive load on the auditory system, allowing the brain to treat conflicting intervals as spatially distinct rather than as a single, harsh blend. Unlike traditional voice‑leading techniques that rearrange note order or alter voicings, dichotic harmony preserves the original score and merely re‑routes notes to different ears, offering composers a low‑cost, reversible tool for enhancing listener comfort. Limitations include dependence on headphone use, individual variations in ear dominance or auditory sensitivity, and the current restriction to a two‑channel (stereo) configuration. The authors suggest extensions to multi‑channel surround formats, latency‑optimized implementations for live performance, and adaptive algorithms tailored to listeners with hearing impairments.

In conclusion, the study demonstrates that applying dichotic listening principles to chord construction can meaningfully improve the aesthetic experience of harmonic music. The proposed algorithm and software prototypes provide a practical foundation for further research and for integration into digital audio workstations, music education platforms, and therapeutic sound design. Future work will explore broader channel configurations, real‑time adaptive grouping based on physiological feedback, and cross‑cultural evaluations of harmonic perception under dichotic conditions.