Title :
Feature selection through orthogonal expansion in isolated word recognition
Author :
Lleida, Eduardo ; Nadeu, Climent ; Marino, J.B.
Author_Institution :
Dpto. Teoria de la Senal y Commun., Univ. Politecnica de Cataluna, Barcelona, Spain
Abstract :
The use of an orthogonal expansion for feature selection and data compression in isolated word recognition is presented. Assuming that the spectral evolution, given by a LPC analysis, is a noisy measure in the sense that it is a linear combination of a set of real features, the objective of the orthogonal expansion is to find these real features. The new feature vectors are used to perform the recognition process. The recognition results as well as the meaning of the orthogonal expansion are discussed
Keywords :
data compression; speech recognition; LPC analysis; data compression; feature selection; feature vectors; isolated word recognition; linear predictive coding; orthogonal expansion; spectral evolution; speech recognition; Ambient intelligence; Linear predictive coding; Pattern recognition; Sampling methods; Signal processing; Speech processing; Speech recognition; Testing; Vector quantization; Vocabulary;
Conference_Titel :
Electrotechnical Conference, 1989. Proceedings. 'Integrating Research, Industry and Education in Energy and Communication Engineering', MELECON '89., Mediterranean
Conference_Location :
Lisbon
DOI :
10.1109/MELCON.1989.50030
