Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userRobust Speaker Clustering using Mixtures of von Mises-Fisher Distributions for Naturalistic Audio Streams
59
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Speaker Diarization (i.e. determining who spoke and when?) for multi-speaker naturalistic interactions such as Peer-Led Team Learning (PLTL) sessions is a challenging task. In this study, we propose robust speaker clustering based on mixture of multivariate von Mises-Fisher distributions. Our diarization pipeline has two stages: (i) ground-truth segmentation; (ii) proposed speaker clustering. The ground-truth speech activity information is used for extracting i-Vectors from each speechsegment. We post-process the i-Vectors with principal component analysis for dimension reduction followed by lengthnormalization. Normalized i-Vectors are high-dimensional unit vectors possessing discriminative directional characteristics. We model the normalized i-Vectors with a mixture model consisting of multivariate von Mises-Fisher distributions. K-means clustering with cosine distance is chosen as baseline approach. The evaluation data is derived from: (i) CRSS-PLTL corpus; and (ii) three-meetings subset of AMI corpus. The CRSSPLTL data contain audio recordings of PLTL sessions which is student-led STEM education paradigm. Proposed approach is consistently better than baseline leading to upto 44.48% and 53.68% relative improvements for PLTL and AMI corpus, respectively. Index Terms: Speaker clustering, von Mises-Fisher distribution, Peer-led team learning, i-Vector, Naturalistic Audio.
rate research
Read More
Speaker diarization determines who spoke and when? in an audio stream. In this study, we propose a model-based approach for robust speaker clustering using i-vectors. The ivectors extracted from different segments of same speaker are correlated. We model this correlation with a Markov Random Field (MRF) network. Leveraging the advancements in MRF modeling, we used Toeplitz Inverse Covariance (TIC) matrix to represent the MRF correlation network for each speaker. This approaches captures the sequential structure of i-vectors (or equivalent speaker turns) belonging to same speaker in an audio stream. A variant of standard Expectation Maximization (EM) algorithm is adopted for deriving closed-form solution using dynamic programming (DP) and the alternating direction method of multiplier (ADMM). Our diarization system has four steps: (1) ground-truth segmentation; (2) i-vector extraction; (3) post-processing (mean subtraction, principal component analysis, and length-normalization) ; and (4) proposed speaker clustering. We employ cosine K-means and movMF speaker clustering as baseline approaches. Our evaluation data is derived from: (i) CRSS-PLTL corpus, and (ii) two meetings subset of the AMI corpus. Relative reduction in diarization error rate (DER) for CRSS-PLTL corpus is 43.22% using the proposed advancements as compared to baseline. For AMI meetings IS1000a and IS1003b, relative DER reduction is 29.37% and 9.21%, respectively.
Speaker clustering is the task of forming speaker-specific groups based on a set of utterances. In this paper, we address this task by using Dominant Sets (DS). DS is a graph-based clustering algorithm with interesting properties that fits well to our problem and has never been applied before to speaker clustering. We report on a comprehensive set of experiments on the TIMIT dataset against standard clustering techniques and specific speaker clustering methods. Moreover, we compare performances under different features by using ones learned via deep neural network directly on TIMIT and other ones extracted from a pre-trained VGGVox net. To asses the stability, we perform a sensitivity analysis on the free parameters of our method, showing that performance is stable under parameter changes. The extensive experimentation carried out confirms the validity of the proposed method, reporting state-of-the-art results under three different standard metrics. We also report reference baseline results for speaker clustering on the entire TIMIT dataset for the first time.
The von Mises-Fisher distribution is one of the most widely used probability distributions to describe directional data. Finite mixtures of von Mises-Fisher distributions have found numerous applications. However, the likelihood function for the finite mixture of von Mises-Fisher distributions is unbounded and consequently the maximum likelihood estimation is not well defined. To address the problem of likelihood degeneracy, we consider a penalized maximum likelihood approach whereby a penalty function is incorporated. We prove strong consistency of the resulting estimator. An Expectation-Maximization algorithm for the penalized likelihood function is developed and simulation studies are performed to examine its performance.
Robust estimation of location and concentration parameters for the von Mises-Fisher distribution is discussed. A key reparametrisation is achieved by expressing the two parameters as one vector on the Euclidean space. With this representation, we first show that maximum likelihood estimator for the von Mises-Fisher distribution is not robust in some situations. Then we propose two families of robust estimators which can be derived as minimisers of two density power divergences. The presented families enable us to estimate both location and concentration parameters simultaneously. Some properties of the estimators are explored. Simple iterative algorithms are suggested to find the estimates numerically. A comparison with the existing robust estimators is given as well as discussion on difference and similarity between the two proposed estimators. A simulation study is made to evaluate finite sample performance of the estimators. We consider a sea star dataset and discuss the selection of the tuning parameters and outlier detection.
Large performance degradation is often observed for speaker ver-ification systems when applied to a new domain dataset. Givenan unlabeled target-domain dataset, unsupervised domain adaptation(UDA) methods, which usually leverage adversarial training strate-gies, are commonly used to bridge the performance gap caused bythe domain mismatch. However, such adversarial training strategyonly uses the distribution information of target domain data and cannot ensure the performance improvement on the target domain. Inthis paper, we incorporate self-supervised learning strategy to the un-supervised domain adaptation system and proposed a self-supervisedlearning based domain adaptation approach (SSDA). Compared tothe traditional UDA method, the new SSDA training strategy canfully leverage the potential label information from target domainand adapt the speaker discrimination ability from source domainsimultaneously. We evaluated the proposed approach on the Vox-Celeb (labeled source domain) and CnCeleb (unlabeled target do-main) datasets, and the best SSDA system obtains 10.2% Equal ErrorRate (EER) on the CnCeleb dataset without using any speaker labelson CnCeleb, which also can achieve the state-of-the-art results onthis corpus.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
