Sciweavers

482 search results - page 31 / 97
» icassp 2010
Sort
View
ICASSP
2010
IEEE
13 years 11 months ago
Noise robust exemplar-based connected digit recognition
This paper proposes a noise robust exemplar-based speech recognition system where noisy speech is modeled as a linear combination of a set of speech and noise exemplars. The metho...
Jort F. Gemmeke, Tuomas Virtanen
ICASSP
2010
IEEE
13 years 9 months ago
Feature extraction for robust speech recognition based on maximizing the sharpness of the power distribution and on power floori
This paper presents a new robust feature extraction algorithm based on a modified approach to power bias subtraction combined with applying a threshold to the power spectral dens...
Chanwoo Kim, Richard M. Stern
ICASSP
2010
IEEE
13 years 9 months ago
Clap your hands! Calibrating spectral subtraction for dereverberation
Reverberation effects as observed by room microphones severely degrade the performance of automatic speech recognition systems. We investigate the use of dereverberation by spectr...
Uwe Zah, Korbinian Riedhammer, Tobias Bocklet, Elm...
ICASSP
2010
IEEE
13 years 9 months ago
A sparse component model of source signals and its application to blind source separation
In this paper, we propose a new method of blind source separation (BSS) for music signals. Our method has the following characteristics: 1) the method is a combination of the spar...
Yu Kitano, Hirokazu Kameoka, Yosuke Izumi, Nobutak...
ICASSP
2010
IEEE
13 years 11 months ago
Voice quality evaluation of various codecs
In this paper a large amount of absolute mean opinion scores obtained within a single listening test are presented. Naïve listeners’ preference on different speech signal prope...
Anssi Rämö