A computational auditory scene analysis system for speech segregation and robust speech recognition

Keyphrases

• End-to-end Speech Recognition 100%
• Computational Auditory Scene Analysis 100%
• Speech Segregation 100%
• Time-frequency Units 100%
• Signal of Interest 50%
• System Yield 50%
• Local Time 50%
• Human Listeners 50%
• Automatic Speech Recognition 50%
• Frame-based 50%
• Automatic Speech Recognizer 50%
• Target Speech 50%
• Baseline Performance 50%
• Auditory Scene Analysis 50%
• Multiple Sound Sources 50%
• Harmonicity 50%
• Speech Separation 50%
• Competing Speech 50%
• Uncertainty Decoding 50%
• Speaker Characteristics 50%
• Binary Time-frequency Mask 50%
• Multipitch Tracking 50%

Engineering

• Systems Analysis 100%
• Frame Time 100%
• Robust Speech Recognition 100%
• Sound Source 50%
• Local Time 50%
• Recognizer 50%
• Automatic Speech Recognition 50%
• Individual Source 50%

Computer Science

• Analysis System 100%
• Auditory Scene Analysis 100%
• Robust Speech Recognition 100%
• Speech Recognition 33%
• Source Individual 33%
• Speech Recognizer 33%