The AMI Meeting Transcription System: Progress and Performance

2009 IEEE International Conference on Acoustics, Speech and Signal Processing

2009

In the meeting case scenario, audio is often recorded using Multiple Distance Microphones (MDM) in a non-intrusive manner. Typically a beamforming is performed in order to obtain a single enhanced signal out of the multiple channels. This paper investigates the use of mutual information for selecting the channel subset that produces the lowest error in a diarization system. Conventional systems perform channel selection on the basis of signal properties such as SNR, cross correlation. In this paper, we propose the use of a mutual information measure that is directly related to the objective function of the diarization system. The proposed algorithms are evaluated on the NIST RT 06 eval dataset. Channel selection improves the speaker error by 1.1% absolute (6.5% relative) w.r.t. the use of all channels.

Section: Experiments and Resultsmentioning

confidence: 99%

Mutual information based channel selection for speaker diarization of meetings data

2009 IEEE International Conference on Acoustics, Speech and Signal Processing

2009

“…Speech/non-speech segmentation is obtained using a forced alignment of the reference transcripts on close talking microphone data using the AMI RT06s first pass ASR models [8]. Results are scored against manual references force aligned by an ASR system.…”

Section: Experiments and Resultsmentioning

confidence: 99%

Combination of agglomerative and sequential clustering for speaker diarization

2008 IEEE International Conference on Acoustics, Speech and Signal Processing

2008

Abstract. This paper aims at investigating the use of sequential clustering for speaker diarization. Conventional diarization systems are based on parametric models and agglomerative clustering. In our previous work we proposed a non-parametric method based on the agglomerative Information Bottleneck for very fast diarization. Here we consider the combination of sequential and agglomerative clustering for avoiding local maxima of the objective function and for purification. Experiments are run on the RT06 eval data. Sequential Clustering with oracle model selection can reduce the speaker error by 10% w.r.t. agglomerative clustering. When the model selection is based on Normalized Mutual Information criterion, a relative improvement of 5% is obtained using a combination of agglomerative and sequential clustering.

Section: Experiments and Resultsmentioning

confidence: 99%

Agglomerative information bottleneck for speaker diarization of meetings data

2007 IEEE Workshop on Automatic Speech Recognition &Amp; Understanding (ASRU)

2007

Abstract. In this paper, we investigate the use of agglomerative Information Bottleneck (aIB) clustering for the speaker diarization task of meetings data. In contrary to the state-of-the-art diarization systems that models individual speakers with Gaussian Mixture Models, the proposed algorithm is completely non parametric . Both clustering and model selection issues of nonparametric models are addressed in this work. The proposed algorithm is evaluated on meeting data on the RT06 evaluation data set. The system is able to achieve Diarization Error Rates comparable to state-of-the-art systems at a much lower computational complexity.2 IDIAP-RR 07-31