A New Adaptation Method for Speaker-Model Creation in High-Level Speaker Verification

Abstract

Abstract. Research has shown that speaker verification based on highlevel speaker features requires long enrollment utterances to be reliable. However, in practical speaker verification, it is common to model speakers based a limited amount of enrollment data. To minimize the undesirable effect of insufficient enrollment data on system performance, this paper proposes a new adaptation method for creating speaker models based on high-level features. Different from conventional methods, the proposed adaptation method not only adapts the phoneme-dependent background model but also the phoneme-independent speaker model. The amount of adaptation in the latter is adjusted by a proportional factor derived from the phoneme-independent background models. The proposed method was compared with traditional MAP adaptation under the NIST2000 SRE framework. Experimental results show that the proposed method can solve the data-spareness problem effectively and achieves a better performance when compare with traditional MAP adaptation.

Articulatory-feature based sequence kernel for high-level speaker verification

“…And the MAP adaptation described in [10] is applied to create the adapted speaker models (2) where, β c ∈ [0, 1] is a phonetic class-dependent adaptation coefficient controlling the contribution of the speaker data and the background models (Eq. 1) on the MAP-adapted model.…”

A new adaptation approach to high-level speaker-model creation in speaker verification

“…However, the client models that they created are essentially a linear weighted sum of enrollment data's distribution and background models. It was found that the modeling capability of the AFCPMs drops rapidly when the amount of enrollment data decreases [36,37].…”