Four-part choral synthesis system for investigating intonation in a cappella choral singing

Abstract

Accurate tuning is an important aspect of singing in harmony in the context of a choir or vocal ensemble. Tuning and ‘pitch drift’ are concerning factors in performance for even the most accomplished professional choirs when singing a cappella (unaccompanied). In less experienced choirs tuning often lacks precision, typically because individual singers have not developed appropriate listening skills. In order to investigate accuracy of tuning in ensemble singing situations, a chorally appropriate reference is required against which frequency measurements can be made. Since most basic choral singing involves chords in four parts, a four-part reference template is used in which the fundamental frequencies of the notes in each chord can be accurately set. This template can now be used in experiments where three of the reference parts are tuned in any musical temperament (tuning system), in this case equal and just temperaments, and played over headphones to a singer to allow her/his tuning strategy to be investigated. This paper describes a practical implementation of a four-part choral synthesis system in Pure Data (Pd) and its use in an investigation of tuning of notes by individual singers using an exercise originally written to explore pitch drift in a cappella choral singing.

Key words::

• Choral synthesis
• equal temperament
• intonation
• in-tune singing
• just intonation
• pure data
• singing
• synthesis

Acknowledgements

The authors thank the singers who took part in the experiments and Mark Hildred and Ben Johnson from Apollo Creative for their input and support on implementing the interfaces for the CreST exhibitions.

Declaration of interest: The authors confirm that they have no financial or other conflicts of interests in relation to the work reported herein.

The authors are members of the CreST (Creative Speech Technology) network, and they acknowledge and are grateful for its funding by the UK Engineering and Physical Sciences Research Council (EPSRC grant number EP/I013512/1).

Notes

Notes

Notes

¹These public exhibitions were created as part of the CreST (Creative Speech Technology) network funded by the UK Engineering and Physical Sciences Research Council (EPSRC) in York, Sheffield and Hull (December 2012) and Scarborough (January 2013).