Topic
Dynamic time warping
About: Dynamic time warping is a research topic. Over the lifetime, 6013 publications have been published within this topic receiving 133130 citations.
Papers published on a yearly basis
Papers
More filters
TL;DR: In this article, a modification of the dynamic time warping (DTW) algorithm was proposed to warping spectral batch data, which takes into account the amount of warping information of every process variable.
85 citations
TL;DR: This work shows that the newly introduced multidimensional DTW concept requires significantly less decoding time while providing the same data fusion flexibility as the AHMM, and can be applied in a wide range of real-time multimodal classification tasks.
84 citations
TL;DR: The two methods for on-line signature verification are compared following the protocol defined in the Signature Verification Competition 2004 and fusion results are provided demonstrating the complementary nature of these two approaches.
Abstract: Function-based methods for on-line signature verification are studied. These methods are classified into local and regional depending on the features used for matching. One representative method of each class is selected from the literature. The selected local and regional methods are based on Dynamic Time Warping and Hidden Markov Models, respectively. Some improvements are presented for the local method aimed at strengthening the performance against skilled forgeries. The two methods are compared following the protocol defined in the Signature Verification Competition 2004. Fusion results are also provided demonstrating the complementary nature of these two approaches.
84 citations
01 Oct 1995
TL;DR: It is concluded that digit zero is the best digit for speaker discrimination, and it is shown that there is a large variation in performance across the different digits.
Abstract: The authors evaluate continuous density hidden Markov models (CDHMM), dynamic time warping (DTW) and distortion-based vector quantisation (VQ) for speaker recognition, emphasising the performance of each model structure across incremental amounts of training data. Text-independent (TI) experiments are performed with VQ and CDHMMs, and text-dependent (TD) experiments are performed with DTW, VQ and CDHMMs. For TI speaker recognition, VQ performs better than an equivalent CDHMM with one training version, but is outperformed by CDHMM when trained with ten training versions. For TD experiments, DTW outperforms VQ and CDHMMs for sparse amounts of training data, but with more data the performance of each model is indistinguishable. The performance of the TD procedures is consistently superior to TI, which is attributed to subdividing the speaker recognition problem into smaller speaker-word problems. It is also shown that there is a large variation in performance across the different digits, and it is concluded that digit zero is the best digit for speaker discrimination.
83 citations
06 Sep 2015
TL;DR: An architecture for the unsupervised discovery of talker-invariant subword embeddings using a dynamic-time warping based spoken term discovery system and a Siamese deep neural network.
Abstract: We report on an architecture for the unsupervised discovery of talker-invariant subword embeddings. It is made out of two components: a dynamic-time warping based spoken term discovery (STD) system and a Siamese deep neural network (DNN). The STD system clusters word-sized repeated fragments in the acoustic streams while the DNN is trained to minimize the distance between time aligned frames of tokens of the same cluster, and maximize the distance between tokens of different clusters. We use additional side information regarding the average duration of phonemic units, as well as talker identity tags. For evaluation we use the datasets and metrics of the Zero Resource Speech Challenge. The model shows improvement over the baseline in subword unit modeling.
83 citations