Abstract
The goal of keyword spotting is to detect “keywords” from conversational speech in a speaker-independent mode, while ignoring all other words. The primary difference between keyword spotting (KWS) and CSR is that CSR operates with a closed-set vocabulary and KWS is an open-set problem. A closed-set vocabulary restricts the talker to a pre-defined vocabulary, while an open-set vocabulary implies that the talker can say anything. In addition, KWS typically operates with uncooperative talkers in noisy environments. As such, KWS is particularly sensitive to words, word combinations, or word boundaries which are acoustically similar to a keyword. For example, the keyword “Middleton” might be confused, or “false alarmed,” with middle bin, middle one or even little dent. However, when one attempts to reduce the false alarm rate, it is usually at the expense of introducing missed detections. Thus, there is an inherent performance tradeoff between recognition accuracy and false alarm rate.
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.
