Advances in unsupervised audio classification and segmentation for the broadcast news and NGSW corpora
Authors:
- 18 April 2006
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Audio, Speech, and Language Processing
- Vol. 14 (3) , 907-919
Abstract
The problem of unsupervised audio classification and segmentation continues to be a challenging research problem which significantly impacts automatic speech recognition (ASR) and spoken document retrieval (SDR) performance. This paper addresses novel advances in 1) audio classification for speech recognition and 2) audio segmentation for unsupervised multispeaker change detection. A new algorithm is proposed for audio classification, which is based on weighted GMM Networks (WGN). Two new extended-time features: variance of the spectrum flux (VSF) and variance of the zero-crossing rate (VZCR) are used to preclassify the audio and supply weights to the output probabilities of the GMM networks. The classification is then implemented using weighted GMM networks. Since historically there have been no features specifically designed for audio segmentation, we evaluate 16 potential features including three new proposed features: perceptual minimum variance distortionless response (PMVDR), smoothed zero-crossing rate (SZCR), and filterbank log energy coefficients (FBLC) in 14 noisy environments to determine the best robust features on the average across these conditions. Next, a new distance metric, T/sup 2/-mean, is proposed which is intended to improve segmentation for short segment turns (i.e., 1-5 s). A new false alarm compensation procedure is implemented, which can compensate the false alarm rate significantly with little cost to the miss rate. Evaluations on a standard data set-Defense Advanced Research Projects Agency (DARPA) Hub4 Broadcast News 1997 evaluation data-show that the WGN classification algorithm achieves over a 50% improvement versus the GMM network baseline algorithm, and the proposed compound segmentation algorithm achieves 23%-10% improvement in all metrics versus the baseline Mel-frequency cepstral coefficients (MFCC) and traditional Bayesian information criterion (BIC) algorithm. The new classification and segmentation algorithms also obtain very satisfactory results on the more diverse and challenging National Gallery of the Spoken Word (NGSW) corpus.Keywords
This publication has 28 references indexed in Scilit:
- CU-Move: Advanced In-Vehicle Speech Systems for Route Navigation2005
- Efficient audio stream segmentation via the combined T/sup 2/ statistic and Bayesian information criterionIEEE Transactions on Speech and Audio Processing, 2005
- Multifeature audio segmentation for browsing and annotation2003
- The LIMSI Broadcast News transcription systemSpeech Communication, 2002
- Audio content analysis for online audiovisual data segmentation and classificationIEEE Transactions on Speech and Audio Processing, 2001
- Progress in Broadcast News transcription at Dragon Systems1999
- Content-based classification, search, and retrieval of audioIEEE MultiMedia, 1996
- Robust feature-estimation and objective quality assessment for noisy speech recognition using the Credit Card corpusIEEE Transactions on Speech and Audio Processing, 1995
- Principal Component Analysis1986
- Distance measures for speech processingIEEE Transactions on Acoustics, Speech, and Signal Processing, 1976