A NOVEL DTW-BASED DISTANCE MEASURE FOR SPEAKER SEGMENTATION
- 1 January 2006
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2006 IEEE Spoken Language Technology Workshop
Abstract
We present a novel distance measure for comparing two speech segments that uses a local version of the well-known DTW algorithm. Our approach is based on the idea of finding word-level speech patterns that are repeated by the same speaker. Using this distance measure, we develop a speaker segmentation procedure and apply it to the task of segmenting multi-speaker lectures. We demonstrate that our approach is able to generate segmentations that correlate well to independently generated human segmentations. In experiments performed on over ten hours of multi-speaker lecture data, we were able to find speaker change points with precision and recall rates of 80% and 100%, respectively.Keywords
This publication has 5 references indexed in Scilit:
- Speaker Detection Without ModelsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2006
- Scale-space filtering: A new approach to multi-scale descriptionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Towards unsupervised pattern discovery in speechPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Efficient algorithms for locating the length-constrained heaviest segments with applications to biomolecular sequence analysisJournal of Computer and System Sciences, 2002
- DISTBIC: A speaker-based segmentation for audio data indexingSpeech Communication, 2000