Wakeword Audio Buffering for Low-Power Speech Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speech capture devices face challenges in balancing power consumption with operational efficiency, particularly in standby modes where they need to listen for wake words while minimizing power usage.
Innovation Solution
Implementing a dual processor system where a digital signal processor performs sound detection in a low power mode, transitioning the central processing unit into an active mode upon detecting a wake word, and employing duty cycling for components like Wi-Fi to alternate between active and low power modes based on functionality requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If the device operates in low power mode with wake word detection, then power consumption is reduced, but response time and detection accuracy may be compromised
Solution Approach 1:
The system is divided into two distinct processing paths: a low-power digital signal processor (DSP) that continuously monitors for wake words, and a high-power central processing unit (CPU) that handles full speech processing. This segmentation allows the device to maintain accurate wake word detection while consuming minimal power during idle periods, resolving the contradiction between power savings and detection reliability.
Solution Approach 2:
The DSP acts as an intermediary component between the microphone and the CPU. It continuously processes audio input in low-power mode and only activates the CPU when a wake word is detected. This intermediary approach ensures reliable wake word detection without requiring the high-power CPU to remain active, thus maintaining detection accuracy while minimizing power consumption.
2Reliability
If the device uses continuous audio monitoring, then wake word detection reliability is improved, but power consumption increases
Solution Approach 1:
The audio processing function is segmented between two processors with different power characteristics. The DSP handles continuous low-power monitoring, while the CPU remains dormant until needed. This segmentation enables reliable continuous monitoring without the high power consumption that would result from using the full-power CPU continuously.
Solution Approach 2:
The system changes the operational parameters of the processing unit based on the current state. During idle periods, the DSP operates at low power with simplified wake word detection algorithms. When a wake word is detected, the system transitions to high-power mode with full speech processing capabilities. This dynamic parameter adjustment maintains detection reliability while optimizing power consumption.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
A system configured to perform low power mode wakeword detection is provided. A device reduces power consumption without compromising functionality by placing a primary processor into a low power mode and using a secondary processor to monitor for sound detection. The secondary processor stores input audio data in a buffer component while performing sound detection on the input audio data. If the secondary processor detects a sound, the secondary processor sends an interrupt signal to the primary processor, causing the primary processor to enter an active mode. While in the active mode, the primary processor performs wakeword detection using the buffered audio data. To reduce a latency, the primary processor processes the buffered audio data at an accelerated rate. In some examples, the device may further reduce power consumption by including a second buffer component and only processing the input audio data after detecting a sound.