http://cs.uef.fi/sipu/pub/SRE-review-Kinnunen-and-Li.pdf

An overview of text-independent speaker recognition: From features to supervectors

A comparison of PCA, KPCA and ICA for dimensionality reduction in support vector machine

As financial time series are inherently noisy and non-stationary, it is regarded as one of the most challenging applications of time series forecasting. Due to the advantages of generalization capability in obtaining a unique solution, support vector regression (SVR) has also been successfully applied in financial time series forecasting. In the modeling of financial time series using SVR, one of the key problems is the inherent high noise. Thus, detecting and removing the noise are important but difficult tasks when building an SVR forecasting model. To alleviate the influence of noise, a two-stage modeling approach using independent component analysis (ICA) and support vector regression is proposed in financial time series forecasting. ICA is a novel statistical signal processing technique that was originally proposed to find the latent source signals from observed mixture signals without having any prior knowledge of the mixing mechanism. The proposed approach first uses ICA to the forecasting variables for generating the independent components (ICs). After identifying and removing the ICs containing the noise, the rest of the ICs are then used to reconstruct the forecasting variables which contain less noise and served as the input variables of the SVR forecasting model. In order to evaluate the performance of the proposed approach, the Nikkei 225 opening index and TAIEX closing index are used as illustrative examples. Experimental results show that the proposed model outperforms the SVR model with non-filtered forecasting variables and a random walk model.

Financial time series forecasting using independent component analysis and support vector regression

A review of biometric technology along with trends and prospects

It is demonstrated that the distribution of speech samples is well described by Laplacian distribution (LD). The widely known speech distributions, i.e., LD, Gaussian distribution (GD), generalized GD, and gamma distribution, are tested as four hypotheses, and it is proved that speech samples during voice activity intervals are Laplacian random variables. A decorrelation transformation is then applied to speech samples to approximate their multivariate distribution. To do this, speech is decomposed using an adaptive Karhunen-Loeve transform or a discrete cosine transform. Then, the distributions of speech components in decorrelated domains are investigated. Experimental evaluations prove that the statistics of speech signals are like a multivariate LD. All marginal distributions of speech are accurately described by LD in decorrelated domains. While the energies of speech components are time-varying, their distribution shape remains Laplacian.

Speech probability distribution

Learning statistically efficient features for speaker recognition

We apply independent component analysis for extracting an optimal basis to the problem of finding efficient features for a speaker. The basis functions learned by the algorithm are oriented and localized in both space and frequency, bearing a resemblance to Gabor functions. The speech segments are assumed to be generated by a linear combination of the basis functions, thus the distribution of speech segments of a speaker is modeled by a basis, which is calculated so that each component should be independent upon others on the given training data. The speaker distribution is modeled by the basis functions. To assess the efficiency of the basis functions, we performed speaker classification experiments and compared our results with the conventional Fourier-basis. Our results show that the proposed method is more efficient than the conventional Fourier-based features, in that they can obtain a higher classification rate.

/pdf/learning-statistically-efficient-features-for-speaker-2i6llhwvft.pdf

This paper presents a feature parameter transformation method using ICA (independent component analysis) for text independent speaker identification of telephone speech ICA is a signal processing technique which can separate linearly mixed signals into statistically independent signals The proposed method transforms them into new vectors using ICA assuming that the cepstrum vectors of the telephone speech collected from various kinds of channel conditions are linear combinations of some characteristic functions with random noise added The performance of the proposed method was compared to the original cepstrum for the HMM-based speaker identification system Experiments were done in equal and different channel conditions on SPIDRE, a real telephone speech database for text independent speaker identification The identification rates increased from about 1 13% most cases, so it was confirmed that the proposed method is effective for speaker identification systems, and more effective in adverse environments

Feature vector transformation using independent component analysis and its application to speaker identification.

Performance Improvement of Speaker Recongnition System for Small Training Data

Speech segregation based on sub-band masking extracts speech signals from audio mixtures via estimation of speech pitch and conservation of signals compatible with the estimated pitch. As speech and music exhibit similar harmonic structures, speech pitch and music pitch coexist in the estimated pitch when speech-music mixture is used as the input, which leads to performance degradation. In order to overcome this limitation, we propose pitch classification using bidirectional 224 정보과학회 컴퓨팅의 실제 논문지 제 25 권 제 4 호(2019. 4) LSTM. The probabilistic attention layer is also proposed to improve the bidirectional LSTM. Further, musical energy removal for segregation mask generation is also proposed in order to obtain naturally segregated speech with pitch classification. The experiment results show that the proposed pitch classification using bidirectional LSTM based on probabilistic attention outscores other speech segregation methods.

/pdf/pitch-classification-based-on-bidirectional-lstm-with-3iuvapfp9c.pdf

Yung-Hwan Oh

Papers

Learning statistically efficient features for speaker recognition

Learning statistically efficient features for speaker recognition

Feature vector transformation using independent component analysis and its application to speaker identification.

Performance Improvement of Speaker Recongnition System for Small Training Data

Pitch Classification Based on Bidirectional LSTM with Probabilistic Attention for Speech Segregation from Speech-Music Mixtures