Topic segmentation on spoken documents using self-validated acoustic cuts

Chen, Hongjie; Xie, Lei; Feng, Wei; Zheng, Lilei; Zhang, Yanning

doi:10.1007/s00500-014-1383-9

Cited by 7 publications

(1 citation statement)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Acoustic pattern matching can be implemented in various variants of dynamic time warping (DTW) which is used to align two sequences of acoustic feature vectors in various tasks, e.g. speech pattern discovery [1,2], speech summarization [26], story segmentation [27,28], etc. In this paper, we employ subsequence-DTW (SDTW) described in [18] to conduct acoustic pattern matching between a query and a test utterance in the retrieval archive.…”

Section: Detection Based On Pattern Matchingmentioning

confidence: 99%

Unsupervised Bottleneck Features for Low-Resource Query-by-Example Spoken Term Detection

et al. 2016

Self Cite

View full text Add to dashboard Cite

We propose a framework which ports Dirichlet Gaussian mixture model (DPGMM) based labels to deep neural network (DNN). The DNN trained using the unsupervised labels is used to extract a low-dimensional unsupervised speech representation, named as unsupervised bottleneck features (uBNFs), which capture considerable information for sound cluster discrimination. We investigate the performance of uBNF in queryby-example spoken term detection (QbE-STD) on the TIMIT English speech corpus. Our uBNF performs comparably with the cross-lingual bottleneck features (BNFs) extracted from a DNN trained using 171 hours of transcribed telephone speech in another language (Mandarin Chinese). With the score fusion of uBNFs and cross-lingual BNFs, we gain about 10% relative improvement in terms of mean average precision (MAP) comparing with the cross-lingual BNFs. We also study the performance of the framework with different input features and different lengths of temporal context.

show abstract