MLLR Techniques for Speaker Recognition

Marc Ferràs (1) , Cheung Chi Leung (1) , Claude Barras (1) , Jean-Luc Gauvain (1)

1 LIMSI - Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur

Maximum-Likelihood Linear Regression (MLLR) and Constrained MLLR (CMLLR) have been recently used for feature extraction in speaker recognition. These systems use (C)MLLR transforms as features that are modeled with Support Vector Machines (SVM). This paper evaluates and compares several of these approaches for the NIST Speaker Recognition task. Single CMLLR and up to 4-phonetic-class MLLR transforms are explored using Gaussian Mixture Models (GMM) and large-vocabulary speech recognition Hidden Markov Models (HMM), using both speaker recognition and speech recognition cepstral front-ends and normalizations. Results for the individual systems as well as in combination with two standard cep-stral systems are provided. Relative gains of 3% and 12% were obtained when combining the best performing CMLLR-based and MLLR-based systems with two standard cepstral systems, respectively.

Domaines: Informatique [cs]; Traitement du signal et de l'image [eess.SP]

Marc Ferràs, Cheung Chi Leung, Claude Barras, Jean-Luc Gauvain. MLLR Techniques for Speaker Recognition. Odyssey 2008: The Speaker and Language Recognition Workshop, Jan 2008, Stellenbosch, South Africa. ⟨hal-01690275⟩