Two educational tools developed from history of speech communication research: Digital pattern playback and vocal-tract models

Digital Pattern Playback [Arai et al., Acoust. Sci. Technol., 27(6), 393-395, (2006)], or DPP, was inspired by the Pattern Playback, originally developed at Haskins Laboratories in the late 1940s. It not only converts original spectrograms of utterances back into sounds, but we can extract and/or modify important cues from it and then resynthesize speech sounds. This approach has allowed us to test the locus theory and many other hypotheses and helped us to develop the area of speech science. In this computer era, DPP is now able to test theories and still teach us more. Another tool originally from speech research that we often use is a set of vocal-tract models. Historically, physical models were used to test whether measured and simplified configurations of the human vocal tract can actually produce target vowels. It turned out such models are very intuitive and useful for seeing the basics of speech production, including vowels and consonants. Thus, like these two tools, knowing more about the history of speech communication research facilitates our future studies. [Work supported by JSPS KAKENHI Grant No. 24K06423.]