Towards Natural Acoustic Interfaces for Automatic Speech Recgnition

Kellermann W (2009)


Publication Language: English

Publication Type: Conference contribution

Publication year: 2009

City/Town: St. Petersburg, Russia

Conference Proceedings Title: Proc. 13th Int. Conf. Speech and Computer

Event location: St. Petersburg RU

Abstract

Aiming at 'natural' hands-free acoustic human/machine inter-faces, the need for according distant-talking automatic speech recognition (ASR) systems increases and presents us with ma-jor signal processing challenges at the acoustic front-end. Con-sidering interactive TV as a challenging exemplary application scenario, we investigate the structural problems presented by noisy and reverberant multi-source environments with unpre-dictable interference and acoustic echoes of loudspeaker sig-nals, and discuss current acoustic signal processing techniques to enhance the input to the actual ASR system. For illustration, the scenario of an EU-funded project on distant-talking inter-faces for interactive TV (DICIT) is considered as a challenging example.

Authors with CRIS profile

How to cite

APA:

Kellermann, W. (2009). Towards Natural Acoustic Interfaces for Automatic Speech Recgnition. In Proc. 13th Int. Conf. Speech and Computer. St. Petersburg, RU: St. Petersburg, Russia.

MLA:

Kellermann, Walter. "Towards Natural Acoustic Interfaces for Automatic Speech Recgnition." Proceedings of the 13th Int. Conf. Speech and Computer, St. Petersburg St. Petersburg, Russia, 2009.

BibTeX: Download