Peripheral Preprocessing In Hearing And Psychoacoustics As Guidelines For Speech Recognition

Authors

  • Eberhard Zwicker

Abstract

Modem electronic equif'llent realizi network of system-theory as  well as signal-theory strategies was a strong motor within the last  15 years  pushing speech recognition systems to better  and better results (for summaries see for example DeMori, 1979; Terhardt, 1978). Nevertheless, this progress is not comparable with the much larger progress of the data processing system like computers, memories, signal processors. Therefore we may ask for other and better guidelines to organize speech recognition systema. since the human hearing system is still by far the best speech recognition system in  every  respect,  it may be  very helpful  to simulate this system   as  much as we know about it. This idea is not  new.  Our  re­ search group seems to offer proposals in this direc­ tion each seventh year (Zwicker, 1971;  Zwicker  et al., 1979), this paper included. Other groups have accepted this aR?roach in part by using critical band filtering CKlatt, 1982), by using loudness-time func­ tions for segmentation !Mermelstein, 1975; Schotola, 1984), or more in general by using loudness-critical band rate-time patterns as preprocessed data base (Ruske,   1985  and  this  vol ume►.

Additional Files

Published

1986-07-21

How to Cite

1.
Zwicker E. Peripheral Preprocessing In Hearing And Psychoacoustics As Guidelines For Speech Recognition. Canadian Acoustics [Internet]. 1986 Jul. 21 [cited 2024 Apr. 18];14(3 bis):1-4. Available from: https://jcaa.caa-aca.ca/index.php/jcaa/article/view/3494

Issue

Section

Proceedings of the Acoustics Week in Canada