Measuring and Adjusting Intonation in Choir Recordings

Sebastian Rosenzweig, Simon Schwär


A central challenge for a cappella singers is to adjust their intonation and to stay in tune relative to their fellow singers. During the editing of a cappella recordings, a sound engineer may want to adjust local intonation of individual singers or compensate for global intonation drifts over time. In this talk, we present methods for measuring and adjusting intonation in recordings of polyphonic singing. First, we introduce a technique for applying time-varying pitch shifts to audio recordings based on resampling and non-linear time-scale modification (TSM). Second, we present a differentiable cost measure for intonation processing that combines tonal and harmonic aspects. We demonstrate the potential of our methods for adjusting intonation in a cappella performances and provide publicly available implementations.

Bio Sebastian:

Sebastian Rosenzweig received his B.Sc. degree in Mediatechnology from Ilmenau University of Technology in 2015 and his M.Sc. degree in Communications and Multimedia Engineering at Friedrich-Alexander-Universität Erlangen-Nürnberg in 2017. He currently is a Ph.D. candidate at the International Audio Laboratories Erlangen under the supervision of Prof. Dr. Meinard Müller. His research interests include audio signal processing and machine learning in the context of polyphonic vocal music.

Bio Simon:

Simon Schwär received his Bachelor’s degree in Audio Engineering from Robert Schumann Conservatory and University of Applied Sciences in Düsseldorf, Germany. He joined Fraunhofer IIS in 2017, where he is developing novel computational methods for VR/AR audio rendering. In 2022, he obtained a Master’s degree in Advanced Signal Processing and Communications Engineering from Friedrich-Alexander Universität Erlangen-Nürnberg. His primary research interests are music information retrieval and virtual acoustics.