A supplementary feature for speaker verification in constrained room acoustics

Abstract

The paper describes a preliminary experiment to find a supplementary feature for speaker verification. In conventional speaker verification systems, spectral features, such as the Mel frequency cepstral coefficient (MFCC), are used universally for all speakers. For some specific applications demanding high security, however, the system also needs to adopt speaker specific supplementary features. Assuming that verification is done in a specified room environment, we analyze the effect of the room response on speaker verification. Simulation results show that the additional supplementary feature is crucial for improving system performance.

Related Papers

• → Infant cry recognition based on feature extraction(2010)3 cited
• → Pitch-based cepstral features for gender classification in noisy environments(2013)1 cited
• A Robust Mel-frequency Cepstrum Coefficients(2008)
• Application of Biomimetic Technology to Feature Extraction from Acoustic Objects(2014)
• → A supplementary feature for speaker verification in constrained room acoustics(2005)