Dynamic Time Warping-based Korean Spoken Word Detection System Using Euclidean Distance in Intelligent Personal Assistants
Dong‐Hyun Lee, Ho-Sung Park, Minkyu Lim, Ji‐Hwan Kim · 2019
This paper proposes a Dynamic Time Warping (DTW)-based Korean spoken word detection system using Euclidean distance (ED) in intelligent personal assistants. The proposed system searches a spoken word from users in a reference utterance and generates a recognized word. For searching, DTW algorithm with ED is used. The accuracy of the proposed system was measured by 91.42% in terms of word error rate (WER).