Abstract
Lip-smack is one of the background events that current speech recognizers can not effectively handle. Little work has been done on the detection of lip-smack. In this paper, lip-smack characteristics in spontaneous speech are analyzed, and then a lip-smack detection method is proposed. A continuous signal is first split into silence and non-silence signal segments, and then potential lip-smack segments are extracted from the non-silence signal segments. Then hidden Markov models (HMMs) are trained and employed to recognize lip-smacks from potential lip-smack segments. The experimental results evaluated on three Mandarin spontaneous speech corpora show that the proposed method can effectively detect lip-smacks in spontaneous speech with an average precision rate and recall rate of 80.9% and 87.4%, respectively.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
