ANALYSIS OF DATA BALANCING PROBLEM IN ACOUSTIC MODELING OF AUTOMATIC SPEECH RECOGNITION SYSTEM
Annotation
The problem of data balancing for training of acoustic models for automatic speech recognition system is considered. A metric is proposed which enables an explicit account for the data level in a cluster during triphone clustering. The proposed approach is shown to improve the quality of speech recognition.
Keywords
Постоянный URL
Articles in current issue
- ANALYSIS OF LANGUAGE STATISTICAL ASPECTS AND THEIR GENDER VARIATIONS BY THE EXAMPLE OF LITHUANIAN
- DETERMINATION OF CHANNEL-INDEPENDENT INFORMATION INDICATORS
- ANALYSIS OF DATA BALANCING PROBLEM IN ACOUSTIC MODELING OF AUTOMATIC SPEECH RECOGNITION SYSTEM
- CROSS-VALIDATION STATE CONTROL IN ACOUSTIC MODEL TRAINING OF AUTOMATIC SPEECH RECOGNITION SYSTEM
- STATISTICAL METHODS FOR AUTOMATIC PROSODIC BREAK DETECTION IN A TEXT-TO-SPEECH SYSTEM
- TIME DELAY ESTIMATION OF AUDIO SIGNALS USING THEIR ENVELOPES
- SPEECH SIGNALS STOCHASTICITY AND ITS EVALUATION
- ASSESSMENT OF FUNCTIONAL SAFETY OF DETECTION OF VIBROACOUSTIC SIGNAL FROM ARRIVING TRAIN WITH ENERGY SENSOR
- TARGET AND NON-TARGET SPEECH SEPARATION USING A DUAL MICROPHONE SYSTEM
- EXPERT SYSTEMS AND METHODS FOR SPEAKER IDENTIFICATION
- CONCEPT OF THE NATIONAL VOICE ACCOUNTING AND VOICE BIOMETRIC SEARCH SYSTEM
- ANALYSIS OF MANIFOLD LEARNING METHODS APPLICABILITY TO SPEAKER RECOGNITION
- EMPLOYMENT OF DTW-BASED HMM-GMM MULTI-SESSION TRAINING IN TEXTDEPENDENT SPEAKER VERIFICATION
- STUDY OF VOICE VERIFICATION SYSTEM TOLERANCE TO SPOOFING ATTACKS USING A TEXT-TO-SPEECH SYSTEM