Self-Defined Text-Dependent Wake-Up-Words Speaker Recognition System

Keyphrases

• Consumer Applications 20%
• CPU Utilization 20%
• Gaussian Mixture Model 40%
• Hidden Markov Model 40%
• Levenshtein Distance 20%
• Mel Frequency Cepstral Coefficients 20%
• Memory Consumption 20%
• Self-defined 100%
• Signal-to-noise Ratio 20%
• Speaker Recognition System 100%
• Speech Features 40%
• Speech Input 40%
• State Sequences 20%
• Text-dependent 100%
• Training Phase 40%
• Two-stage Training 20%
• Voice Activity Detection 20%
• Voice Characteristics 20%
• Wake-up 100%
• Whole Systems 20%

Computer Science

• Average Accuracy 33%
• Claimed Identity 33%
• Consumer Application 33%
• Experimental Result 33%
• Gaussian Mixture Model 66%
• Levenshtein Distance 33%
• Mel Frequency Cepstral Coefficient 33%
• noise signal 33%
• Preprocessing 33%
• Recognition System 100%
• Speaker Recognition 100%
• Training Phase 66%
• Voice Activity Detection 33%