Rapid speaker adaptation in eigenvoice space

Abstract

This paper describes a new model-based speaker adaptation algorithm called the eigenvoice approach. The approach constrains the adapted model to be a linear combination of a small number of basis vectors obtained offline from a set of reference speakers, and thus greatly reduces the number of free parameters to be estimated from adaptation data. These "eigenvoice" basis vectors are orthogonal to each other and guaranteed to represent the most important components of variation between the reference speakers. Experimental results for a small-vocabulary task (letter recognition) given in the paper show that the approach yields major improvements in performance for tiny amounts of adaptation data. For instance, we obtained 16% relative improvement in error rate with one letter of supervised adaptation data, and 26% relative improvement with four letters of supervised adaptation data. After a comparison of the eigenvoice approach with other speaker adaptation algorithms, the paper concludes with a discussion of future work.

A statistical approach for person verification using human behavioral patterns

“…Adaptation techniques such as MAP [24], MLLR [25], and eigenspacebased techniques [26] are often used to solve this problem. Although eigenspace-based techniques are effective when adaptation data are extremely small [38], they restrict the model to a lower dimensionality where much information might be lost [39]. On the other hand, MLLR and MAP do not impose this restriction on the models.…”

Statistical Parametric Speech Synthesis

“…However, if we increase the number of representative HMM sets to enhance the capabilities of representation, it is difficult to determine the interpolation ratio to obtain the required voice. To address this problem, Shichiri et al applied the eigenvoice technique (Kuhn et al, 2000) to HMM-based speech synthesis (Shichiri et al, 2002). A speaker-specific "super-vector" was composed by concatenating the mean vectors of all state-output distributions in the model set for each S speaker-dependent HMM set.…”

Acoustic Model Training Using Pseudo-Speaker Features Generated by MLLR Transformations for Robust Speaker-Independent Speech Recognition

“…This method achieves efficient adaptation. Adaptation techniques which only need a small amount of target speech data, such as those using inter-speaker variation modeling like Eigenvoice [5], have also been proposed. In this framework, the super vectors of the mean parameters of the speaker-dependent acoustic models are used as bases, and the super vector of the new speakerspecific acoustic models is expressed as a linear combination of these bases.…”