Strategies to Improve the Robustness of Agglomerative Hierarchical Clustering Under Data Source Variation for Speaker Diarization

Han, Ki Jin; Kim, S.; Narayanan, Shrikanth

doi:10.1109/tasl.2008.2002085

Cited by 52 publications

(40 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to group similar speech utterances together, we performed a single Gaussian based bottom-up agglomerative hierarchical clustering (AHC) proposed by Han et al (2008) with K-means post refinement, using generalized likelihood ratio (GLR) distance. The smoothness constraint was then enforced by an ad-hoc scheme of jointly classifying all utterances inside a cluster using a majority voting rule.…”

Section: Joint Classification Of Test Samplesmentioning

confidence: 99%

Automatic intelligibility classification of sentence-level pathological speech

Kim¹,

Kumar²,

Tsiartas³

et al. 2015

Computer Speech & Language

View full text Add to dashboard Cite

Pathological speech usually refers to the condition of speech distortion resulting from atypicalities in voice and/or in the articulatory mechanisms owing to disease, illness or other physical or biological insult to the production system. Although automatic evaluation of speech intelligibility and quality could come in handy in these scenarios to assist experts in diagnosis and treatment design, the many sources and types of variability often make it a very challenging computational processing problem. In this work we propose novel sentence-level features to capture abnormal variation in the prosodic, voice quality and pronunciation aspects in pathological speech. In addition, we propose a post-classification posterior smoothing scheme which refines the posterior of a test sample based on the posteriors of other test samples. Finally, we perform feature-level fusions and subsystem decision fusion for arriving at a final intelligibility decision. The performances are tested on two pathological speech datasets, the NKI CCRT Speech Corpus (advanced head and neck cancer) and the TORGO database (cerebral palsy or amyotrophic lateral sclerosis), by evaluating classification accuracy without overlapping subjects’ data among training and test partitions. Results show that the feature sets of each of the voice quality subsystem, prosodic subsystem, and pronunciation subsystem, offer significant discriminating power for binary intelligibility classification. We observe that the proposed posterior smoothing in the acoustic space can further reduce classification errors. The smoothed posterior score fusion of subsystems shows the best classification performance (73.5% for unweighted, and 72.8% for weighted, average recalls of the binary classes).

show abstract

Section: Joint Classification Of Test Samplesmentioning

confidence: 99%

Automatic intelligibility classification of sentence-level pathological speech

Kim¹,

Kumar²,

Tsiartas³

et al. 2015

Computer Speech & Language

View full text Add to dashboard Cite

show abstract

“…The bottom-up approach is the most popular and has achieved general success in the NIST RT evaluations [13]- [17] and on evaluation data in other domains [18], [19]. Nonetheless, some authors report that instabilities related to initialization [20], model merging and the sensitivity of the stopping criterion [21] might degrade its performance.…”

Section: A Bottom-upmentioning

confidence: 99%

“…In our experience progressive training leads to significant improvements in performance over a conventional AHC system. Cluster merging is controlled with the modified Information Change Rate (ICR) criterion [21] and continues until the stopping criterion is met. In contrast to the exemplary, state-of-the-art system presented in [28], we find that the stopping criterion gives better results than the criterion as used in [28] and [29].…”

Section: B Bottom-upmentioning

confidence: 99%

A Comparative Study of Bottom-Up and Top-Down Approaches to Speaker Diarization

Evans

Bozonnet

Wang

et al. 2012

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-This paper presents a theoretical framework to analyze the relative merits of the two most general, dominant approaches to speaker diarization involving bottom-up and top-down hierarchical clustering. We present an original qualitative comparison which argues how the two approaches are likely to exhibit different behavior in speaker inventory optimization and model training: bottom-up approaches will capture comparatively purer models and will thus be more sensitive to nuisance variation such as that related to the speech content; top-down approaches, in contrast, will produce less discriminative speaker models but, importantly, models which are potentially better normalized against nuisance variation. We report experiments conducted on two standard, single-channel NIST RT evaluation datasets which validate our hypotheses. Results show that competitive performance can be achieved with both bottom-up and top-down approaches (average DERs of 21% and 22%), and that neither approach is superior. Speaker purification, which aims to improve speaker discrimination, gives more consistent improvements with the top-down system than with the bottom-up system (average DERs of 19% and 25%), thereby confirming that the top-down system is less discriminative and that the bottom-up system is less stable. Finally, we report a new combination strategy that exploits the merits of the two approaches. Combination delivers an average DER of 17% and confirms the intrinsic complementary of the two approaches.

show abstract

“…In [13] oracle experiments were used to analyze existing and new stopping criteria. Both the speech/non-speech classification and speaker change detection components were replaced by oracle components and only the clustering and stopping components were tested.…”

Section: A Oracle Based Experiments In Other Studiesmentioning

confidence: 99%

Speaker Diarization Error Analysis Using Oracle Components

Huijbregts

Leeuwen

Wooters³

2012

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-In this paper we describe an analysis of our speaker diarization system based on a series of oracle experiments. In this analysis, each system component is substituted by an oracle component that uses the reference transcripts to perform flawlessly. By placing the original components back into the system one at a time, either in a top-down or bottom-up manner, the performance of each individual system component is measured. The analysis approach can be applied to any speaker diarization system that consists of a concatenation of separate components. Our experimental findings are relevant for most RT09s diarization systems that all apply similar techniques. The analysis revealed that three components caused most errors: speech activity detection, the inability to handle overlapping speech and robustness of the merging component to cluster impurity.

show abstract

Strategies to Improve the Robustness of Agglomerative Hierarchical Clustering Under Data Source Variation for Speaker Diarization

Cited by 52 publications

References 16 publications

Automatic intelligibility classification of sentence-level pathological speech

Automatic intelligibility classification of sentence-level pathological speech

A Comparative Study of Bottom-Up and Top-Down Approaches to Speaker Diarization

Speaker Diarization Error Analysis Using Oracle Components

Contact Info

Product

Resources

About