Audio proto objects for improved sound localization

In this article we present a new framework for auditory processing that combines feature extraction and grouping processes to form what we call audio proto objects. These proto objects combine an arbitrary number of audio features in a compact representation that allows a more precise sound localization and also better interfacing to behavior-control in robotics. We compare our standard sound localization system with the new approach in several scenarios to demonstrate the potential of the new approach.

Authors

• Tobias Rodemann
• Frank Joublin
• Christian Goerick

Keywords

• Speech recognition
• Signal processing
• Streaming media
• Time measurement
• Acoustic noise
• Filters
• Histograms
• Intelligent robots
• USA Councils
• Feature extraction
• Sound Localization
• Short-term Memory
• Integration Time
• Segment Length
• Localization Performance
• C=O Groups
• Position Estimation
• Signal Energy
• Segmentation Process
• Scalar Product
• Correct Estimation
• Sound Source
• Humanoid Robot
• Formant
• Temporal Integration
• Sound Files
• Population Coding
• Feature Compression
• Audio Stream
• Auditory Scene Analysis
• Background Noise
• Localization Error
• Time Constant
• Environmentally Friendly
• Direct Quotes
• Azimuth Angle
• Secondary Loss
• Position Vector
• System Architecture