Sciweavers

123 search results - page 9 / 25
» A Geometric Approach for Separating Several Speech Signals
Sort
View
SPEECH
2008
97views more  SPEECH 2008»
13 years 7 months ago
A new approach for the adaptation of HMMs to reverberation and background noise
Looking at practical application scenarios of speech recognition systems several distortion effects exist that have a major influence on the speech signal and can considerably det...
Hans-Günter Hirsch, Harald Finster
ICASSP
2010
IEEE
13 years 5 months ago
A short-time objective intelligibility measure for time-frequency weighted noisy speech
Existing objective speech-intelligibility measures are suitable for several types of degradation, however, it turns out that they are less appropriate for methods where noisy spee...
Cees H. Taal, Richard C. Hendriks, Richard Heusden...
TASLP
2010
138views more  TASLP 2010»
13 years 2 months ago
Glimpsing IVA: A Framework for Overcomplete/Complete/Undercomplete Convolutive Source Separation
Abstract--Independent vector analysis (IVA) is a method for separating convolutedly mixed signals that significantly reduces the occurrence of the well-known permutation problem in...
Alireza Masnadi-Shirazi, Wenyi Zhang, Bhaskar D. R...
ICASSP
2008
IEEE
14 years 2 months ago
Maximum a posteriori ICA: Applying prior knowledge to the separation of acoustic sources
Independent component analysis (ICA) for convolutive mixtures is often applied in the frequency domain due to the desirable decoupling into independent instantaneous mixtures per ...
Graham W. Taylor, Michael L. Seltzer, Alex Acero
IMC
2007
ACM
13 years 9 months ago
Blind source separation approach to performance diagnosis and dependency discovery
We consider the problem of diagnosing performance problems in distributed system and networks given end-to-end performance measurements provided by test transactions, or probes. C...
Gaurav Chandalia, Irina Rish