Source filter theory is central in the modern description of speech production. The key concepts underpinning the theory are the sustained oscillation of vocal folds and the vocal tract resonances that modulate the source signal. In this short essay the development of these concepts is superficially followed from early 17th to late 19th century. Marin Mersenne was the first to experimentally verify the relationship between the pitch of a sound (of a cord) and its fundamental frequency, published in 1636. By analogy, Mersenne concluded that vocal folds must vibrate with the same frequency as a string with the same pitch. While Mersenne could not explain why several frequencies could be simultaneously present in glottal excitation, Daniel Bernoulli suggested superposition of waves as a solution in 1753, which was further developed by Joseph Fourier. Johannes Müller’s report of excised larynx experiments in 1839 can be considered as the first version of source filter theory. Finally, once X-ray imaging during speech production was made possible in 1897, the stage was set for experimental validation of source filter theory.
Cite as: Aalto, D. (2017) Origins of source filter theory from Mersenne (1636) to Müller (1839). Proc. Second International Workshop on the History of Speech Communication Research (HSCR 2017), 61-67, doi: 10.21437/HSCR.2017-7
@inproceedings{aalto17_hscr,
author={Daniel Aalto},
title={{Origins of source filter theory from Mersenne (1636) to Müller (1839)}},
year=2017,
booktitle={Proc. Second International Workshop on the History of Speech Communication Research (HSCR 2017)},
pages={61--67},
doi={10.21437/HSCR.2017-7}
}