ISCA & IEEE Workshop on Spontaneous Speech Processing and Recognition April 13-16, 2003 Tokyo Institute of Technology, Tokyo, Japan

Group Delay Based Segmentation of Spontaneous Speech into Syllable-Like Units

T. Nagarajan, Hema A. Murthy, Rajesh M. Hegde

Department of Computer Science and Engineering, Indian Institute of Technology, Madras, India

In the development of a syllable-centric ASR system, segmentation of the acoustic signal into syllabic units is an important stage. This paper presents a minimum phase group delay based approach to segment spontaneous speech into syllable-like units. Here, three different minimum phase signals are derived from the short term energy functions of three sub-bands of speech signals, as if it were a magnitude spectrum. The experiments are carried out on Switchboard corpus and the error in segmentation is found to be utmost 40msec for 85% of the syllable segments, in addition to 5.25% insertions and 7.10% deletions.

Full Paper

Bibliographic reference.  Nagarajan, T. / Murthy, Hema A. / Hegde, Rajesh M. (2003): "Group delay based segmentation of spontaneous speech into syllable-like units", in SSPR-2003, paper MAP20.