An investigation of choral blending through soundfield capture, acoustic evaluation, and perceptual analysis methods
Jithin Thilakan and Ying-Ying Zhang have completed a preliminary study in Detmold on their Collaborative Student Project Grant, "An investigation of choral blending through soundfield capture, acoustic evaluation, and perceptual analysis methods." They will carry out further research in Montreal this summer.
The aim of their project is to evaluate musical blend in choral performance from both signal and perceptual analysis methods. They will analyze choral blend on two levels: at the sound source level and the room acoustic level. The influence of feedback from the acoustic environment—along with coordinated performance strategies between musicians—is observed to have a high impact on individual musical performances and significantly affect the resulting sound field of joint performance. In order to evaluate the source-level blending of individual voices in realistic choir performances, recording the individual singers with minimal microphone cross-talk interference and room reflections is a challenge.
As a preliminary investigation, the evaluation of microphone cross-talk between singers in different acoustic environments was carried out in Detmold. Two singers (one male, one female) were recorded to evaluate cross-talk between signals recorded in the same space but under different acoustic settings. Recordings utilized an extremely close lavalier microphone technique as well as a more traditional large diaphragm capture. Signal analysis and critical listening results from this study will help determine an appropriate method of choosing and positioning microphones and sources for recording performers that balances signal leakage with realistic musical performance.
An expanded methodology will be utilized in additional experiments in Montreal involving small ensembles of singers performing in unison or in harmony. The goal of these experiments is to test additional acoustic, musician-feedback, and microphone capture parameters to find the least amount of signal leakage while also allowing singers a naturalistic performance and blend. Parameters that present the least amount of leakage balanced with musical performance will be utilized to conduct signal and perceptual evaluations on choral blending.