Wearable Song Identification via Vocal Trigger
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wearable computing devices with near-eye displays, such as head-mountable devices (HMDs), face challenges in efficiently identifying and processing songs played in the environment without draining the battery, as they require significant processing power and may provide more information than desired, such as a complete list of all songs heard throughout the day.
Innovation Solution
A wearable computing device configured to receive sensor data from sensors indicating humming, singing, or whistling by the wearer, triggering a song identification algorithm to recognize and display only the songs the wearer enjoys, reducing processing demands and battery drain by selectively initiating content recognition based on user engagement signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the wearable computing device continuously performs content recognition on all ambient audio, then complete song identification is achieved, but processing power requirements and battery consumption increase significantly
Solution Approach 1:
The system performs preliminary analysis by detecting humming, singing, or whistling patterns before initiating full content recognition. This preliminary action filters out non-relevant audio periods, preventing unnecessary processing and battery consumption while maintaining complete song identification when the wearer is actively engaged with the music
Solution Approach 2:
Instead of continuously performing full content recognition, the system applies partial action by only triggering detailed analysis when specific conditions (humming, singing, whistling detection) are met. This selective approach reduces overall processing requirements while ensuring complete identification of songs that matter to the wearer
2Loss of information
If the wearable computing device processes all audio data throughout the day, then all songs are identified, but the wearer receives more information than desired
Solution Approach 1:
The system uses feedback from sensor data (detecting humming, singing, or whistling patterns) to dynamically control when content recognition should be performed. This feedback mechanism automatically filters out periods when the wearer is not engaged with music, eliminating the need for complex manual information filtering while maintaining completeness of desired song information
Solution Approach 2:
The system serves itself by using its own sensor data to automatically determine when song identification should occur. The wearable device monitors the wearer's behavior patterns and autonomously decides when to activate content recognition, eliminating the need for external control or complex filtering algorithms while providing precisely the information the wearer wants
3Productivity
If the wearable computing device runs song identification algorithm continuously, then no songs are missed, but processing power is wasted during periods of no user engagement
Solution Approach 1:
The system transitions from continuous periodic action to conditional periodic action by triggering content recognition only during periods when humming, singing, or whistling is detected. This maintains song identification throughput for engaged periods while eliminating processing energy waste during non-engaged periods, as the algorithm runs periodically only when necessary
Data Source
AI summary
The present disclosure provides a wearable computing device. The wearable computing device may include a control system configured to perform functions. The functions may include receiving sensor data from one or more sensors of the wearable computing device. The functions may also include determining whether the sensor data is indicative of humming, singing, or whistling by a wearer. The functions may also include causing the wearable computing device to perform a content recognition of audio content in an ambient environment of the wearable computing device in response to the sensor data being indicative of humming, singing, or whistling by the wearer.