Application Processor Voice Trigger System Low Power
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice-based intelligent interfaces require tactile input to initiate voice recognition, which consumes power and detracts from the user experience, especially in devices that rely on limited battery resources.
Innovation Solution
An application processor with an integrated voice trigger system that performs voice trigger operations using a trigger input signal, reducing power consumption by offloading tasks from the host processor and enabling voice recognition without tactile input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If continuous or intermittent monitoring of audio channel is performed to detect voice input, then voice trigger operation can be initiated, but electrical power consumption increases
Solution Approach 1:
The patent divides the audio processing system into separate functional blocks: a voice trigger system for detecting trigger events and a host processor for general operations. The voice trigger system includes dedicated components (audio interface, voice trigger processor, memory) that operate independently to monitor audio channels and detect wake words, allowing the host processor to remain in low-power states while voice triggering functionality is maintained
Solution Approach 2:
The patent introduces a dedicated voice trigger processor as an intermediary component between the audio interface and the host processor. This intermediate processor handles the computationally intensive task of voice recognition and trigger event detection, freeing the main host processor from continuous audio monitoring duties and enabling it to enter power-saving modes
2Ease of operation
If voice-based intelligent interface is implemented without tactile input, then hands-free operation is achieved, but power consumption increases due to continuous monitoring
Solution Approach 1:
The voice trigger system performs preliminary action by continuously monitoring audio input and pre-processing voice data to detect trigger events before activating the full voice-based intelligent interface. The system detects wake words or trigger phrases in advance, allowing the host processor to remain dormant until actually needed, thus enabling hands-free operation while minimizing power consumption
3Device complexity
If voice trigger system is integrated in single semiconductor chip with host processor and audio subsystem, then device complexity is reduced, but manufacturing precision requirements increase
Solution Approach 1:
The patent merges the voice trigger system, host processor, and audio subsystem into a single integrated semiconductor chip. The voice trigger system includes dedicated components (audio interface, voice trigger processor, memory) that are physically integrated with the host processor through a system bus, reducing the overall number of discrete components and simplifying the device architecture while maintaining functional independence through careful design of interconnections and power management
Data Source
AI summary
An application processor includes a system bus, a host processor, a voice trigger system and an audio subsystem that are electrically connected to the system bus. The voice trigger system performs a voice trigger operation and issues a trigger event based on a trigger input signal that is provided through a trigger interface. The audio subsystem includes an audio interface and processes audio streams through the audio interface. The voice trigger system is disposed in an always-powered domain where power is supplied in both of an active mode and a standby mode. The host processor and the audio subsystem are disposed in a power-save domain where power is blocked in the standby mode. The host processor launches into the active mode when the voice trigger system issues the trigger event.