On the existence of speaker-specific maximum flow declination rate (MFDR)–sound pressure level (SPL) profiles

Oral flow was registered with a Rothenberg mask in two male subjects during the utterance of an /æ/ vowel embedded in a word at three intensity levels (soft, normal, and loud voice), and during the production of crescendos and decrescendos on the same vowel at three fundamental frequencies. Registered signals were digitized and processed with specific hard- and software [J. S. Perkell, J. Acoust. Soc. Am. 89, 1777–1781 (1991)] to give the MFDR. This is a measure for the decrease of airflow through the glottis, during the closing phase of the glottal cycle. MFDR values showed a speaker-specific relationship with SPL. To explore this specificity, MFDR values were measured in 70 subjects during phonation of the vowel /æ/ with soft, normal, and loud voice. The relation between MFDR and SPL can be described with MFDR=a+bSPL. The dependency of the parameters a and b on gender, voice training, and vocal pathology will be discussed. [Support was given by the Foundation for Linguistic Research, NWO.]