Articulation Index (AI): Concepts and Applications 2000-01-3150

The paper presents the concepts and applications regarding Articulation Index (AI) that is widely used in the automotive industry as a powerful parameter to qualify/quantify the middle and high frequency spectra associated to the internal noise of vehicles. We are able to show a short historical review including the today used definitions and calculations based in third octave band spectra through and idealized speech region model like our pioneer Brazilian standard. We show also other applications, where AI can be an excellent tool to evaluate vehicle acoustics. We emphasize the relation of AI to Speech Intelligibility and make considerations concerning other factors influencing Intelligibility. In addition we comment the integration we have produced with an overall noise parameter like dB(A) - (H-index) and the proposition of an AIM - Articulation Index Modified.

Finally we discuss the results presented and identify the position of AI in the general context/vision of the parameters that are available in vehicle acoustics.