Audio Classification for Portable Device Energy Conservation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Portable computing devices face energy drain and reduced battery life due to constantly running computationally expensive audio-identification processes, even when no media content is being presented by surrounding media devices.

Innovation Solution

Implementing a method where a portable computing device uses a microphone to classify ambient audio as containing media content or not, employing a trained machine-learning model to determine audio properties and statistical measures, thereby controlling when to engage in audio-identification processes, conserving resources by forgoing identification when no media content is detected.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the portable computing device constantly runs audio-identification processes to identify media content, then the accuracy of media content identification is improved, but the energy consumption increases and battery life decreases

Engineering Contradiction:

Improvemedia content identification accuracyVSAvoidenergy consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system performs audio classification periodically to determine whether media content is present, and only engages in audio-identification processes when media content is detected. This periodic classification approach replaces continuous audio-identification processing, significantly reducing energy consumption while maintaining identification accuracy when needed.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The system performs preliminary audio classification before engaging in the more computationally intensive audio-identification process. By pre-screening the audio environment to detect media content presence, the system avoids unnecessary full identification processes and reduces overall energy consumption.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If the portable computing device constantly runs audio-identification processes to identify media content, then the media content identification capability is improved, but the battery life decreases

Engineering Contradiction:

Improvemedia content identification capabilityVSAvoidbattery life

Core Design Contradiction:

ReliabilityVSDuration of action of stationary object

Solution Approach 1:

The system uses periodic audio classification to monitor the environment for media content, triggering audio-identification processes only when necessary. This periodic approach extends battery life by minimizing the duration of high-power processing while maintaining reliable identification capability when media content is present.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The preliminary audio classification step screens the audio environment before initiating full audio-identification processes. This ensures that identification capability is maintained when needed while avoiding unnecessary processing that would drain the battery.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If the portable computing device engages in audio-identification processes whenever audio is present, then the completeness of media content detection is improved, but the computational resource usage increases

Engineering Contradiction:

Improvemedia content detection completenessVSAvoidcomputational resource usage

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the audio processing task into two distinct stages: audio classification (determining media content presence) and audio identification (identifying specific media content). This segmentation allows the system to use lightweight classification continuously while reserving resource-intensive identification for when media content is detected, reducing overall computational resource usage.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Audio classification serves as a preliminary filtering step before full audio identification. This preliminary action ensures that computational resources are only fully utilized when media content is actually present, maintaining detection completeness while reducing unnecessary resource consumption during non-media periods.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250016404A1Use of Audio Classification as Basis to Control Audio Identification

Publication Date: 2025.01.09 THE NIELSEN CO (US) LLC

AI summary

A method includes receiving, into a microphone of a portable computing device, audio from a surrounding environment of the portable computing device. The method also includes classifying, by the portable computing device, the received audio as containing media content or as containing no media content. Classifying the received audio as containing media content or as containing no media content comprises determining whether the audio defines content emitted from a media player in the surrounding environment of the portable computing device. The method further includes, based on the classifying, controlling by the portable computing device whether to engage in an audio-identification process for determining an identity of the media content.