Natural Speech Detection Using Multi-Band Head Orientation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In environments with multiple smart devices, determining the focus of a user's attention and which device a command is directed at is challenging due to the difficulty in distinguishing between microphone signals from different devices.
Innovation Solution
The system utilizes the frequency-dependent nature of human speech propagation to determine a user's orientation relative to smart devices by analyzing power levels in multiple frequency bands of microphone signals, employing spectral characteristics and neural networks to estimate the user's facing direction, and combining this with device-specific data to identify the focus of attention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple smart devices are deployed in a smart environment, then the functionality and coverage of the voice user interface is improved, but the ability to determine which device the user is talking to deteriorates
Solution Approach 1:
The patent introduces a new dimension for device identification by analyzing the spectral characteristics of speech signals. Instead of relying solely on spatial position or device ID, the system examines the frequency-dependent power distribution across multiple bands, creating a spectral fingerprint that links the user's orientation to specific devices.
Solution Approach 2:
The system changes the parameter space by comparing power levels across multiple frequency bands rather than using a single overall signal strength metric. This multi-dimensional spectral analysis (examining ratios and patterns across different frequency ranges) enables precise determination of user orientation and device focus.
2Measurement precision
If spectral analysis across multiple frequency bands is performed, then the precision of user orientation determination is improved, but the computational complexity increases
Solution Approach 1:
The speech signal is segmented into multiple frequency bands, and the analysis is performed independently on each band. This segmentation allows the complex spectral analysis to be broken down into manageable components (power level calculations for each band), which can then be recombined to determine overall orientation.
Solution Approach 2:
Different frequency bands are analyzed with specific attention to their local spectral characteristics. The system examines power ratios and patterns in specific frequency ranges (e.g., low-frequency vs. high-frequency bands) to extract orientation information, applying local quality analysis rather than uniform processing across the entire spectrum.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Accurately determines the user's focus of attention, enabling effective command association and control of smart devices based on the user's orientation, enhancing the functionality of voice user interfaces in smart environments.
Implementation Method 1
The system utilizes the frequency-dependent nature of human speech propagation to determine a user's orientation relative to smart devices by analyzing power levels in multiple frequency bands of microphone signals
Data Source
AI summary
A system, comprising: a plurality of distributed smart devices comprising: a first smart device having a first microphone; a second smart device having a second microphone; and processing circuitry, configured to: receive, from the first microphone, a first microphone signal comprising speech of a user; receive, from the second microphone, a second microphone signal comprising the speech of the user; determine an orientation of the user's head relative to the first smart device and the second smart device based on the first microphone signal and the second microphone signal, wherein determining the orientation of the user's head comprises comparing first power levels in a plurality of frequency bands of the first microphone signal; and controlling one or more of the plurality of distributed smart devices based on the determined orientation.