Multi-Microphone Channel Selection Using Composite Keyword Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional keyword detection techniques require significant computational resources and may incorrectly select multiple channels when the same keyword utterance is detected by multiple microphones, leading to inefficiencies.
Innovation Solution
A channel selection apparatus that combines input voice signals from multiple channels into a composite signal, calculates power for each channel, and selects the channel with the maximum power based on keyword detection results, optionally using candidate selection and weighted power calculations to reduce computational load and maintain signal-to-noise ratio.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If keyword detection processing is performed on each channel separately, then the keyword detection accuracy is maintained, but the computation amount increases significantly
Solution Approach 1:
The patent segments the keyword detection process into two stages: first performing keyword detection on a composite signal formed by adding all channels, then selectively applying further processing only to channels that contain the keyword. This segmentation reduces computation by avoiding full keyword detection processing on all channels when the keyword is present in only one channel.
Solution Approach 2:
The patent merges all channel signals into a single composite signal for initial keyword detection. By combining the signals first and then identifying which channel contains the keyword based on the detection result and power comparison, the system achieves accurate keyword detection with reduced computational load compared to processing each channel independently.
2Reliability
If all channels detecting the keyword are selected, then the keyword detection coverage is complete, but the channel selection accuracy deteriorates when the same keyword utterance is collected by multiple microphones
Solution Approach 1:
The patent applies local quality by evaluating the power level of each individual channel to determine which channel should be selected for output. Instead of treating all channels equally, the system identifies the channel with the maximum power level among those containing the keyword, thereby selecting the channel with the highest signal quality for further processing.
Solution Approach 2:
The patent changes the selection criterion from binary detection (keyword present/absent) to a continuous parameter-based selection (power level comparison). By comparing the power levels of channels containing the keyword and selecting the channel with maximum power, the system achieves more precise channel selection while maintaining complete keyword detection coverage.
Data Source
AI summary
A channel in which an utterance of a keyword is included is selected from acoustic signals of multiple channels. An addition unit 11 adds all channels of input voice signals of multiple channels to generate a composite voice signal of one channel. A keyword detection unit 12 generates a keyword detection result indicating a result of detecting an utterance of a predetermined keyword from a composite voice signal. A power calculation unit 13 calculates powers of channels based on input voice signals. A delay unit 14 delays the powers of the channels. When the keyword detection result indicates that the keyword was detected, a maximum power detection unit 15 selects, as an output channel, a channel having the maximum power among the powers of the channels of the input voice signals. A channel selection unit 16 selects the voice signal of the output channel from the input voice signals and outputs the selected voice signal.