Statistical voice activity detection using low-variance spectrum estimation and an adaptive threshold
MetadataShow full item record
Traditionally voice activity detection algorithms are based on any combination of general speech properties such as temporal energy variations, periodicity, and spectrum. This paper describes a novel statistical method for voice activity detection using a signal to noise ratio measure. The method employs a low-variance spectrum estimate and determines an optimal threshold based on the estimated noise statistics. A possible implementation is presented and evaluated over a large test set and compared to current modern standardized algorithms. The evaluations indicate promising results with the proposed scheme being comparable or favorable over the whole test set.
Showing items related by title, author, creator and subject.
Kühnapfel, Thorsten (2009)For humans, hearing is the second most important sense, after sight. Therefore, acoustic information greatly contributes to observing and analysing an area of interest. For this reason combining audio and video cues for ...
Effective online learning experiences: exploring potential relationships between Voice-over-Internet-Protocol (VoIP) learning environments and adult learners’ motivation, multiple intelligences, and learning stylesScott, Donald E. (2009)This study was a 360 degree exploration of the effectiveness of online learning experiences facilitated via Voice-over-Internet-Protocol (VoIP) by incorporating the insights afforded by students, their lecturers, and the ...
Yong, Pei Chee; Nordholm, Sven; Dam, Hai Huyen; Leung, Yee-Hong; Lai, Chiong Ching (2013)The real-time implementation of the existing multi-channel Wiener filter (MWF) algorithms suffer from performance degradation due to the lack of robustness against estimation errors of the second-order statistics. The ...