Abstract
Currently, various procedures exist for physical measurements of speech intelligibility (AI, STI, RASTI, etc.). A large majority of them are either based on or use some concepts of the articulation theory. In this talk the articulation theory will be critically reviewed and parameters and procedures that give best results will be recommended. The specific recommendations will vary depending on the application. Therefore, the result will be a global and unified model incorporating the best elements of the contemporary procedures. Among the variables that will be discussed are the basic theoretical foundations of the theory, speech dynamic range, importance of various frequencies to speech intelligibility, ways to account for the effects of various distortion types on speech intelligibility, accurate measurement of the prediction parameters, possible errors, etc. Finally, the meaning of the term “speech intelligibility” will be reexamined. It will be hypothesized that the intelligibility of speech that occurs in daily communication activities is better specified by physical measures than by any speech recognition testing. In fact, in many circumstances, the latter could be completely misleading.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.