Journal
Scientific and technical journal «Priborostroenie»
UDK681.3
Issue:2 (56)
Download PDF417 Kbyte
COMPARISON OF VARIOUS MIXTURES OF GAUSSIAN PLDA-MODELS IN THE PROBLEM OF TEXT-INDEPENDENT SPEAKER VERIFICATION
Annotation
Applicability of unsupervised mixtures of PLDA models with Gaussian priors in a i-vector space for speaker verification is studied. Conditions under which the application is advantageous are analyzed for existing training databases. A mixture of two PLDA models is shown to be more effective than a single PLDA model for a cross-channel task.
Keywords
Постоянный URL
Articles in current issue
- A PHONETICALLY RICH TEXT FOR FUNDAMENTAL AND APPLIED RESEARCH ON RUSSIAN SPEECH VARIABILITY
- SUPPRESSION OF ACOUSTIC NOISE IN AUDIO DEVICE USING ASYNCHRONOUS REFERENCE SIGNAL
- ALGORITHMS FOR DETECTION OF TYPICAL NOISES AND INTERFERING BURSTS IN SPEECH SIGNALS
- MODERN MOBILE SYSTEM FOR TRACK WARNING
- AUTOMATION OF NEW VOICE CREATION PROCEDURE FOR A RUSSIAN TTS SYSTEM
- A HYBRID TECHNOLOGY FOR TTS SYSTEM BASED ON HIDDEN MARKOV MODELS AND UNIT SELECTION ALGORITHM
- ASSESSMENT OF SYNTHESIZED SPEECH QUALITY: PROBLEMS AND SOLUTIONS
- APPLICATION OF LINGUISTIC ANALYSIS FOR TEXT NORMALIZATION AND HOMONYMY RESOLUTION IN RUSSIAN TEXT-TO-SPEECH SYSTEM
- STUDY OF INFORMATIVE SPEECH FEATURES FOR AUTOMATIC SPEAKER IDENTIFICATION
- GINI CRITERION SVM FOR EMOTION CLASSIFICATION FRAMEWORK
- FEATURES OF HUMAN-MACHINE INTERFACE OF MODERN BIOMETRIC IDENTIFICATION SYSTEMS
- EVALUATION OF THE CONFIDENCE INTERVAL FOR DECISION PREDICTION OF AN ENSEMBLE OF CLASSIFIERS