Title :
On Low-level Cognitive Components of Speech
Author :
Feng, Ling ; Hansen, Lars Kai
Author_Institution :
Intelligent Signal Process., Tech. Univ. Denmark, Lyngby
Abstract :
In this paper we analyze speech for low-level cognitive features using linear component analysis. We demonstrate generalizable component fingerprints´ stemming from both phonemes and speakers. Phonemes are fingerprints found at the basic analysis window time scale (20 msec), while speaker ´voiceprints´ are found at time scales around 1000 msec. The analysis is based on homomorphic filtering features and energy based sparsification
Keywords :
independent component analysis; speaker recognition; speech processing; cognitive component analysis; linear component analysis; phoneme recognition; speaker recognition; speech processing; Humans; Independent component analysis; Informatics; Mathematical model; Principal component analysis; Signal analysis; Signal processing; Signal processing algorithms; Speech analysis; Unsupervised learning;
Conference_Titel :
Computational Intelligence for Modelling, Control and Automation, 2005 and International Conference on Intelligent Agents, Web Technologies and Internet Commerce, International Conference on
Conference_Location :
Vienna
Print_ISBN :
0-7695-2504-0
DOI :
10.1109/CIMCA.2005.1631575
