DocumentCode :
2288832
Title :
On the Discrimination of Speech/Music Using a Time Series Regularity
Author :
Swe, Ei Mon Mon ; Pwint, Moe ; Sattar, Farook
Author_Institution :
Univ. of Comput. Studies, Yangon
fYear :
2008
fDate :
15-17 Dec. 2008
Firstpage :
53
Lastpage :
60
Abstract :
A new method to discriminate between speech and music related to the automatic transcription of broadcast news is presented. In the proposed method, a time series regularity, sample entropy (SampEn), is mainly used as an efficient feature to discriminate speech and music of broadcast audio stream. SampEn is a variant of the approximate entropy (ApEn) that measures the regularity of time series. Depending on the regularity of time series, a segment of a given audio stream is classified into speech or music. The first step of the method is calculation of SampEn sequence over windows. The second step is classification of this segment with a rule-based classification scheme over sample entropy sequence. Experimental results show the effectiveness of the proposed method for broadcast news shows with different music styles.
Keywords :
audio signal processing; audio streaming; broadcasting; entropy; music; signal classification; speech processing; time series; approximate entropy; audio stream broadcasting; audio stream segmentation classification; automatic broadcast news transcription; rule-based classification scheme; sample entropy sequence; speech/music discrimination; time series regularity; Automatic speech recognition; Broadcast technology; Digital multimedia broadcasting; Entropy; Hidden Markov models; Multiple signal classification; Robustness; Streaming media; Time domain analysis; Time measurement;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Multimedia, 2008. ISM 2008. Tenth IEEE International Symposium on
Conference_Location :
Berkeley, CA
Print_ISBN :
978-0-7695-3454-1
Electronic_ISBN :
978-0-7695-3454-1
Type :
conf
DOI :
10.1109/ISM.2008.19
Filename :
4741147
Link To Document :
بازگشت