Sound reinforcement and reproduction systems have grown to a scale where church committees and auditorium owners spend up to $1,000,000 for them. In the past, such systems were purchased without any assurance that acceptable speech intelligibility would be achieved. Today speech intelligibility can be specified in advance, designed for, and objectively measured with an accuracy as good as that achieved using a panel of "live" listeners. The competing techniques are described and evaluated, and some of the remaining problem areas encountered in analyzing nonstandard systems are outlined. the Peutz percent articulation loss of consonants technique (%ALcons), the speech transmission index (STI), and the rapid speech transmission index (RASTI) are all defined, compared to live listener tests made with a large listener sample, and illustrated using currently available analyzers.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.