Birdsong—A quantitative acoustic model

Abstract

Most discussions of birdsong are at the descriptive level or, at best, involve only a qualitative discussion of the acoustical mechanisms involved in sound production. Qualitatively, two different types of song can be distinguished and these may be described as “voiced” or “whistled,” respectively, though these terms do not necessarily refer to the sound-production mechanism. There is general agreement among biologists that the “voiced” song is produced by vibration of the membranes of the vocal organ (the syrinx) under the influence of aerodynamic and pressure forces, though hitherto no quantitative acoustical analysis has been attempted. The present paper puts forward such an analysis using the large body of knowledge that has been built up on the study of reed-blown musical instruments. It is shown that this approach allows a quantitative description to be formulated which yields pressure and flow waveforms and thus the form of the radiated spectrum of the song and a value for the total radiated power. Results are in quite good agreement with observations. The inability of the model to yield the nearly pure-tone sounds characteristic of “whistled” birdsong suggests that this song may be produced by an entirely different mechanism and may be truly whistled.