Re: National differences in the tritone paradox (at)

Subject: Re: National differences in the tritone paradox
From:    at <parncuttSOUND.MUSIC.MCGILL.CA>
Date:    Fri, 17 Dec 1993 15:09:23 GMT-0500

On Pierre Divenyi's recommendation, I am forwarding this letter from John Ohala <ohala(at)> (with his permission). The letter begins with a quote from my second letter to the list of October 30, 1993. > ... It is clear > that the timbre, and hence spectral envelope, of specific vowels > differs from one accent, dialect, or language to another. For > example, an American "oh" is timbrally deeper than an English "oh", > suggesting that the American "oh" has a lower centroid frequency. > Might American vowels be generally (or, at least, on average) > "timbrally deeper" than English ones? Has anyone measured long-time > average spectra of the speech of different languages and dialects to > find out if there are consistent differences in the distributions? First, there is someone who has looked at LTAS for different languages, but I'm not sure he has included English or dialects of English in his survey: Bernard Harmegnies (Av. de Champ de Mars CH II / Univ de L'Etat / B - 7000 Mons / Belgium; Fax +32 65 373054). His interest is voice quality differences between languages. Second, it is misleading to say that Am. Engl. "oh" is "deeper" than a Brit. Engl. "oh"; both are diphthongs, the Brit. Engl. markedly so. ASCII approximations to an IPA transcription would give Am. Engl. version as [ow] and Brit-Engl. as [uh-oo]; since there is no steady-state and both sweep through the same freq. region, I don't see any possibility of saying that one is deeper than the other. Third, in any case, all languages have many vowels which cover much the same space in acoustic-auditory terms. With languages that have front rounded vowels (Fr. [y] as in "tu", [oe] as in "peur") one might think that they would have more frequencies at the high end of F2, but as it happens, these vowels are statistically infrequent in running speech -- so they probably have much the same distribution of resonant frequencies as other languages. In addition the res. freq.'s are constantly changing as a function of consonants; seldom is a given freq. dwelled on. Fourth, the actual resonant frequencies of vowels vary with the size of the vocal tract, being higher for shorter vocal tracts (female vs. male; children vs. adults). You might have a better chance of finding systematically different concentrations of resonant frequencies if you looked at different sexes or ages (but then what is important: the speaker's own res. freq.'s or those of the people they listen to?). One might also find differences as a function of average stature in diff. countries, e.g., Swedes, Iroquoians, and Watusis vs. Andean Quechuas, pygmies. (Assumption: vocal tract length and thus res. freq.'s vary inversely with other bodily dimensions.) The bottom line is that I am extremely skeptical that language differences would lead to significant differences in resonant frequencies. Hope this helps. John

This message came from the mail archive
maintained by:
DAn Ellis <>
Electrical Engineering Dept., Columbia University