Title :
Towards unsupervised pattern discovery in speech
Author :
Park, Alex ; Glass, James R.
Author_Institution :
MIT Comput. Sci. & Artificial Intelligence Lab., Cambridge, MA
Abstract :
We present an unsupervised algorithm for discovering acoustic patterns in speech by finding matching subsequences between pairs of utterances. The approach we describe is, in theory, language and topic independent, and is particularly well suited for processing large amounts of speech from a single speaker. A variation of dynamic time warping (DTW), which we call segmental DTW, is used to performing the pairwise utterance comparison. Using academic lecture data, we describe two potentially useful applications for the segmental DTW output: augmenting speech recognition transcriptions for information retrieval and speech segment clustering for unsupervised word discovery. Some preliminary qualitative results for both experiments are shown and the implications for future work and applications are discussed
Keywords :
information retrieval; pattern clustering; speech recognition; acoustic patterns; dynamic time warping; information retrieval; speech recognition transcriptions; speech segment clustering; unsupervised pattern discovery; unsupervised word discovery; Acoustic testing; Bioinformatics; Genomics; Loudspeakers; Natural languages; Proteins; Sequences; Speech processing; Speech recognition; Training data;
Conference_Titel :
Automatic Speech Recognition and Understanding, 2005 IEEE Workshop on
Conference_Location :
San Juan
Print_ISBN :
0-7803-9478-X
Electronic_ISBN :
0-7803-9479-8
DOI :
10.1109/ASRU.2005.1566529