A recursive expectation-maximization algorithm for speaker tracking and separation

Keyphrases

• Beamformer 25%
• Direction of Arrival Estimation 75%
• Estimation Task 50%
• Frequency Bin 100%
• Linearly Constrained Minimum Variance Beamformer 25%
• Maximization Model 50%
• Mean Absolute Error 25%
• Mixture of Gaussian Model 25%
• Multi-speaker 50%
• Multichannel Wiener Filter 50%
• Multiple Microphones 25%
• Precise Time 25%
• Realistic Scenario 25%
• Recursive Expectation-maximization 100%
• Recursive Expectation-maximization Algorithm 100%
• Relative Transfer Function 25%
• Relative Transfer Function Estimation 50%
• Separation Algorithm 25%
• Separation Capability 25%
• Single-speaker 25%
• Speaker Activity 25%
• Speaker Localization 25%
• Speaker Separation 100%
• Speaker Tracking 100%
• Speech Signal 50%
• Statistical Model 25%
• Time-frequency 75%
• Time-frequency Domain 25%
• Time-frequency Map 25%

Engineering

• Beamformer 40%
• Constrainedness 20%
• Direction-of-Arrival Estimation 60%
• Expectation Maximization Algorithm 100%
• Frequency Bin 80%
• Frequency Domain 20%
• Gaussian Model 20%
• Maximization 80%
• Mean Absolute Error 20%
• Multichannel 40%
• Real Life 20%
• Recursive 100%
• Speech Signal 40%
• Statistical Model 20%
• Transfer Function 60%