Assessing The Performance Of Three Methods For Separating Non-Spontaneous And Spontaneous Speech Through Simulation  

Oleh:

Kenny, Owen P. ; Nelson, Douglas J. ; Bodenschatz, John S. ; McMonagle, Heather

Jenis: Article from Journal - ilmiah internasional
Dalam koleksi: Simulation vol. 70 no. 5 (May 1994), page 304-313.
Topik: Spontaneous speech; non-spontaneous speech; prosodic features; probability distribution; Hellinger’s method; Gaussianity
Fulltext: 304.pdf (743.14KB)

Isi artikel

The ability to distinguish spontaneous from non-spontaneous speech can prove helpful, such as in forensic evidence situations, sorting voice-mail responses from voice-mail menus, and automatic segmentation of spontaneous responses from prepared questions. The latter situation occurs when trying to create a database of spontaneous data from data of a speaker responding spontaneously to prepared prompts. This paper outlines and compares three methods for automatically classifying spontaneous and nonspontaneous speech, and presents the experimental results of the performance of all three methods, evaluated on high quality simulated data. All three methods are based on an analysis of the probability distributions of prosodic features extracted from speech signals.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali