Query-by-example retrieval via fast sequential dynamic time warping algorithm

Abstract

We introduce a novel approach to Query-byExample (QbE) retrieval, utilizing fundamental principles of posteriorgram-based Spoken Term Detection (STD), in this paper. Proposed approach is a kind of modification of widely used segmental variant of dynamic programming algorithm. Our solution represents sequential variant of DTW algorithm, employing one step forward moving strategy. Each DTW search is carried out sequentially, block by block, where each block represents squared input distance matrix, with size equal to the length of retrieved query. We also examine a way how to speed up sequential DTW algorithm without considerable loss in retrieving performance, by implementing linear time-aligned accumulated distance. The increase of detection accuracy is ensured by weighted cumulative distance score parameter. Therefore, we called this approach Weighted Fast Sequential -DTW (WFS-DTW) algorithm. A novel PCA-based silence discriminator is used along with this algorithm. Evaluation of proposed algorithm is carried out on ParDat1 corpus, using Term Weighted Value (TWV).

TEDxSK and JumpSK: A New Slovak Speech Recognition Dedicated Corpus

“…At first level, the silence discrimination is performed by our proposed voice activity detection (VAD) algorithm [16]. In order to determine VAD labels, the waveform is processed in the time domain by overlapping blocks extracted by rectangular window with length of 25ms and 10ms frame step.…”