Voice Activation With DSP Keyword Detection and Host Sleep
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice command processing in devices leads to high power consumption due to continuous monitoring and analysis by the processing circuitry, necessitating a reduction in power usage for efficient operation.
Innovation Solution
Implementing a digital signal processor (DSP) to detect and compare voice instructions with stored keywords, activating a host processor only when a match is found, thereby maintaining the host processor in a sleep state until a keyword is recognized, and using intermittent active and sleep cycles to reduce power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the processing circuitry continuously monitors and analyzes voice input, then voice command detection capability is improved, but power consumption increases dramatically
Solution Approach 1:
The system divides voice processing into two separate stages: first, a low-power keyword spotting stage that continuously monitors for specific trigger words, and second, a full voice command analysis stage that activates only when a keyword is detected. This segmentation allows the device to maintain voice detection capability while dramatically reducing average power consumption by keeping the full processing circuitry in sleep mode most of the time.
Solution Approach 2:
The system performs preliminary keyword detection before activating full voice processing. By continuously monitoring for predefined keywords in a low-power state and only activating the main processing circuitry when a keyword is detected, the system prepares for potential voice commands without maintaining continuous high-power operation, thus resolving the contradiction between detection reliability and power consumption.
2Speed
If the host processor remains in active state to process voice commands, then processing speed is improved, but power consumption increases
Solution Approach 1:
The host processor operates in periodic cycles rather than continuously. It remains in sleep mode during most periods and activates only when a keyword is detected by the low-power monitoring circuitry. This periodic operation maintains the ability to process voice commands at full speed when needed while dramatically reducing average power consumption through extended sleep periods.
Solution Approach 2:
The system performs preliminary keyword detection using a low-power digital signal processor before activating the main host processor. This preliminary action allows the host processor to remain in sleep mode while still being ready to process voice commands at full speed when triggered, thus resolving the contradiction between processing speed and power consumption.
3Measurement precision
If the processing circuitry continuously analyzes voice input, then voice command recognition accuracy is improved, but power consumption increases dramatically
Solution Approach 1:
The voice recognition system is segmented into two functional parts: a continuous low-power keyword spotting function and an intermittent high-accuracy full voice analysis function. The keyword spotting maintains basic recognition accuracy for trigger words while consuming minimal power, and the full analysis provides high accuracy only when needed, thus resolving the contradiction between recognition accuracy and power consumption.
Solution Approach 2:
The system performs partial voice processing continuously by monitoring for keywords, and performs complete voice analysis only when necessary. This partial action approach maintains sufficient recognition accuracy for the keyword trigger while avoiding the excessive power consumption of continuous full-scale voice analysis, effectively resolving the contradiction between accuracy and power usage.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information processing apparatus that detects a voice command via a microphone in order to activate the device and execute certain applications. The apparatus comprises a digital signal processor (DSP) and a host controller which are responsible for processing the voice commands. The DSP recognizes and processes voice commands intermittently while the host processor is in a sleep state, thereby reducing the overall power consumption of the apparatus. Further, when the DSP is configured to recognize voice commands intended only to activate the device, a memory having a sufficiently lower storage capacity suffices.