The ANSI standard method for measurement of the intelligibility of speech over communication systems.
Mones E. Hawley · The Journal of the Acoustical Society of America · 1991
ANSI S3.2-1989 was adopted in December 1989 and replaces the standard adopted in 1961. The new standard is intended for people who want quantified results of comparisons among alternative systems. The standard is not intended for use with untrained subjects, for subjects without normal hearing or speaking abilities, for synthesis-by-rule (text-to-speech) devices, or for automatic speech recognition equipment. The standard is not intended for use to evaluate speaker recognition or identification systems. The standard is the result of many compromises, such as between the needs of military and civilian user groups, between the realism of testing in the field and the reproducibility of laboratory testing, between the desire for economy and the need for scientific validity. The new standard emphasizes the selection and training of talkers and listeners; the design, control, and reporting of test conditions; and the analysis and reporting of test results.