An Automatic Algorithm For Segmenting And Labelling A Connected Digit Sequence

An Automatic Algorithm For Segmenting And Labelling A Connected Digit Sequence (2000)

Cached

Download Links

[speech.iitm.ernet.in]

by V. Kamakshi Prasad , Hema A. Murthy

Citations:

1 - 0 self

Versions

Version History

Metadata Version 2

User correction supplied by SystemCorrections

Datum	Value	Source
TITLE	An Automatic Algorithm For Segmenting And Labelling A Connected Digit Sequence	user correction - Legacy Corrections
AUTHOR NAME	V. Kamakshi Prasad	SVM HeaderParse 0.1
AUTHOR AFFIL	Department of Computer Science and Engineering,; Indian Institute of Technology	SVM HeaderParse 0.2
AUTHOR ADDR	Madras, Chennai- 600 036	SVM HeaderParse 0.1
AUTHOR NAME	Hema A. Murthy	SVM HeaderParse 0.1
AUTHOR AFFIL	Department of Computer Science and Engineering,; Indian Institute of Technology	SVM HeaderParse 0.2
AUTHOR ADDR	Madras, Chennai- 600 036	SVM HeaderParse 0.1
ABSTRACT	Group delay functions provide an alternative representation of signal information. The main features of group delay functions are the additive and high resolution properties. The Fourier transform (FT) phase is generally featureless due to random polority and wrapping. But the group delay function which is defined as the negative derivative of phase, can be processed to derive significant information such as peaks and valleys in the spectral envelope. In this paper, we show an application of group delay function to solve the segmentation problem in speech. In the proposed method a new signal is generated by symmetrising the short term energy function. The minimum phase group delay function of this signal is computed, the valleys of which correspond to segment boundaries. The proposed technique was tested on manually segmented digit utterances of the TI-DIGITS database. The overall correct segmentation performance is 77.8%. Digitwise recognition performance on the correctly segmented database is 87.1%	user correction - Legacy Corrections
YEAR	2000	user correction - Legacy Corrections
CITATIONS	10 found	ParsCit 1.0