Wake Word Detection Circuits for Low-Power Continuous Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques for wake word detection in smart devices are limited by high power consumption and insecure data transmission, and lack efficient methods for identifying wake words in a power-efficient manner.
Innovation Solution
Implementing a dual-circuit system with a low power circuit for initial wake word detection and a high performance circuit for verification, utilizing continuous learning and training to enhance the accuracy and efficiency of wake word detection, where the low power circuit operates at a reduced power level and the high performance circuit verifies and processes queries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If conventional wake word detection techniques are used, then wake word identification can be performed, but power consumption is high
Solution Approach 1:
The system divides wake word detection into two segments: a low power circuit performs initial detection of wake word candidates, and a high performance circuit verifies those candidates. This segmentation allows the majority of operations to run at low power while maintaining high accuracy through selective verification, directly resolving the contradiction between power consumption and detection reliability.
2Measurement precision
If high performance circuit is used for wake word detection, then detection accuracy is improved, but power consumption increases
Solution Approach 1:
The high performance circuit performs verification only on wake word candidates identified by the low power circuit, rather than processing all audio inputs. This partial action approach maintains high detection accuracy where needed while avoiding the excessive power consumption that would result from continuous high performance processing, thus resolving the contradiction between precision and energy use.
3Measurement precision
If continuous learning is implemented, then wake word detection accuracy is improved, but device complexity increases
Solution Approach 1:
A specialized learning circuit acts as an intermediary between the low power and high performance circuits, receiving feedback from the high performance verification results and updating the low power circuit's detection parameters. This intermediary learning mechanism improves accuracy over time while keeping the overall system complexity manageable by confining learning operations to a dedicated component rather than distributing complexity throughout the entire system.
Data Source
AI summary
Systems, methods, and devices detect audio signals. Methods may include receiving an audio input at an audio front end circuit, and identifying, using a first circuit, a wake word based, at least in part, on the received audio signal. Methods may also include determining if the identifying performed by the first circuit should be modified based, at least in part, on a designated threshold of accuracy associated with the first circuit and a result of the identifying, the designated threshold of accuracy being determined based, at least in part, on a language processing model of a second circuit. Moreover, the modifying may include adjusting wake word detection operations of the first circuit based, at least in part, on a result of the determining.