Stimuli

Last updated 28 April 1996.

These are stimuli used in the double-vowel identification experiment.

The experiment's stimulus set contained synthetic steady-state single and double Japanese vowels. For double vowels, there was an F0 difference of 0 or 6%, and a level difference of -20, -10, 0, 10 or 20 dB between the RMS signal levels of the two vowels before mixing. All stimuli were set to the same RMS value before presentation via headphones at a SPL of 63-70dB. Stimulus duration was 200 ms with 20 ms onset and offset ramps.

These are the single vowels in AIFF format, from which all the double vowels can be constructed. They sound terrible via the loudspeaker of my Mac, but all right when I download them back to the NeXT, so I think they are OK (let me know if not!).

/a/ at 125 Hz and 132.5 Hz
/e/ at 125 Hz and 132.5 Hz
/i/ at 125 Hz and 132.5 Hz
/o/ at 125 Hz and 132.5 Hz
/u/ at 125 Hz and 132.5 Hz

The vowels were synthesized by John Culling's 'klatt' program, using formant frequencies from Hirahara, T. and H. Kato (1992). "The effect of F0 on vowel identification" in "Speech perception, production and linguistic structure", Edited by Y. Tohkura, E. Vatikiotis-Bateson and Y. Sagisaka. Tokyo, Ohmsha, 89-112.

Here are some examples of double vowel stimuli for the pair /o/+/u/. The level is of /o/ relative to /u/:

-20 dB at unison and at 6 % delta F0
-10 dB at unison and at 6 % delta F0
0 dB at unison and at 6 % delta F0
10 dB at unison and at 6 % delta F0
20 dB at unison and at 6 % delta F0

Alain's home page and Keele workshop page.