Abstract
In this work, two different keyword search (KWS) methods are proposed in order to improve the existing KWS system which is based on large vocabulary continuous speech recognition (LVCSR) and weighted finite state transducers (WFST). In the first method, a symbolic index is generated by applying vector quantization to the posteriorgram representation of the audio and then WFST based search is performed. In the second method, KWS was done with the subsequence dynamic time warping (sDTW) algorithm which is commonly used in the query-by-example spoken term detection (QbE-STD) tasks. As a result of the experiments, it has been observed that when combined with the existing KWS system, the proposed systems improve the performance especially for the out-of-vocabulary (OOV) queries.
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.
