Rechargeable Casting Device Wake Control for Low-Power Voice Input
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Battery-powered devices face limited charge capacity due to frequent processing of casting requests and user inputs, especially when interacting with automated assistants, leading to reduced available casting time and wastage of computational resources.
Innovation Solution
Implementing a rechargeable device with a digital signal processor (DSP) and a device system on a chip (SoC) that operates in sleep mode to monitor user inputs, adaptively managing wake times based on speech recognition models and offloading processing tasks to subsystems like a WiFi chip to conserve power and resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the device SoC constantly processes casting requests and user inputs, then the automated assistant functionality is responsive and accurate, but the battery charge capacity is depleted quickly
Solution Approach 1:
The patent divides the processing system into two segments: a low-power digital signal processor (DSP) for initial audio monitoring and a high-performance system on chip (SoC) for complex processing. The DSP continuously monitors audio inputs with minimal power consumption, while the SoC is activated only when the DSP detects potential user inputs, thereby segmenting the continuous high-power operation into intermittent high-power bursts.
Solution Approach 2:
Instead of continuous SoC operation, the system implements periodic activation based on detected user inputs. The DSP operates continuously at low power, and the SoC is periodically activated only when needed for processing actual casting requests or assistant commands, converting continuous high-power consumption into periodic low-duty-cycle operation.
2Productivity
If the device SoC processes frequent casting requests, then casting functionality is maintained, but the available casting time is reduced due to frequent processing
Solution Approach 1:
The patent extracts the function of monitoring and initial processing from the main SoC and assigns it to a dedicated low-power DSP. This extraction allows the SoC to focus only on actual casting request processing while the DSP handles continuous monitoring, thereby increasing the proportion of time the SoC can remain in low-power state and extending available casting time.
Solution Approach 2:
The DSP acts as an intermediary between the continuous audio input and the SoC. It filters and pre-processes audio signals, only passing relevant data to the SoC when actual user inputs are detected. This intermediary role reduces the frequency and duration of SoC activation, extending overall casting availability.
3Measurement precision
If the device uses speech recognition models to detect invocation phrases, then automated assistant activation is accurate, but computational resources are wasted when no input is present
Solution Approach 1:
The speech recognition process is segmented into two stages: a lightweight phoneme detection stage performed by the DSP using simplified models, and a full speech recognition stage performed by the SoC only when phonemes are detected. This segmentation ensures high detection accuracy while avoiding full computational resource expenditure during periods of no input.
Solution Approach 2:
The system applies partial speech recognition action through the DSP's phoneme detection capability, which uses simplified models requiring minimal computation. Full speech recognition is applied only when necessary (when phonemes are detected), avoiding excessive computational action during silent periods while maintaining detection accuracy.
Data Source
Figure 1A
Figure 1B
Figure 2A
AI summary
Implementations set forth herein relate to management of casting requests and user inputs at a rechargeable device, which provides access to an automated assistant and is capable of rendering data that is cast from a separate device. Casting requests can be handled by the rechargeable device despite a device SoC of the rechargeable device operating in a sleep mode. Furthermore, spoken utterances provided by a user for invoking the automated assistant can also be adaptively managed by the rechargeable device in order mitigate idle power consumption by the device SoC. Such spoken utterances can be initially processed by a digital signal processor (DSP), and, based on one or more features (e.g., voice characteristic, conformity to a particular invocation phrase, etc.) of the spoken utterance, the device SoC can be initialized for an amount of time that is selected based on the features of the spoken utterance.