User Terminal Volume Control by Sound Scene Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing volume adjustment methods for user terminals, such as mobile phones, fail to accurately match the ring tone and earpiece volumes to the user's current scenario due to incorrect environmental sound analysis, often mistaking human sounds as noise, leading to inappropriate volume settings.
Innovation Solution
A method and apparatus that collect and analyze sound signals to determine the composition of sounds, including blank, human, and noise, to accurately identify the scene mode and adjust the volume accordingly, using volume adjustment coefficients specific to each mode, and dynamically adjust earpiece volume based on real-time microphone input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If volume is adjusted according to decibels of environment sound only, then volume adjustment is simple, but accuracy of volume adjustment is low
Solution Approach 1:
The patent segments the environment sound into different sound types (human sound, noise, blank sound) and analyzes their proportions. Instead of treating environment sound as a single decibel value, the system divides it into categories and determines scene mode based on the composition ratios of these segments, thereby improving accuracy while maintaining operational simplicity.
Solution Approach 2:
The patent changes the parameter basis for volume adjustment from a single decibel value to multiple parameters including sound type composition ratios (human sound proportion, noise proportion, blank sound proportion). This parameter expansion allows the system to distinguish between different sound scenarios more accurately, resolving the contradiction between simplicity and accuracy.
2Extent of automation
If environment sound is used to determine scene mode, then volume adjustment is automated, but scene mode determination accuracy is low
Solution Approach 1:
The patent introduces sound type composition analysis as an intermediary step between environment sound detection and scene mode determination. Instead of directly mapping decibel levels to scene modes, the system first analyzes the sound signal to identify human sounds, noise, and blank sounds, then uses their proportions as intermediate parameters to accurately determine the scene mode, improving automation reliability.
Solution Approach 2:
The patent implements dynamic scene mode determination by continuously analyzing the composition of environment sound in real-time. The system adapts to changing sound scenarios by monitoring the proportions of different sound types and dynamically adjusting the scene mode accordingly, enhancing both automation and accuracy.
3Difficulty of detecting and measuring
If human sound is mistaken for noise, then noise detection sensitivity is high, but scene mode recognition accuracy is low
Solution Approach 1:
The patent applies local quality analysis by examining specific characteristics of sound signals to identify human sounds versus noise. Instead of treating all non-blank sounds uniformly as noise, the system analyzes local features such as frequency patterns and temporal characteristics to distinguish human speech from actual noise, preventing misclassification and improving scene mode recognition accuracy.
Data Source
AI summary
A volume adjustment method comprising performing analysis on the collected sound signal surrounding a user terminal, to obtain composition information, where the composition information includes sound types included in the sound signal and proportions of sounds of the various types, and the sound types include blank sound, human sound, and noise, determining a current scene mode of the user terminal according to the composition information, and adjusting volume of the user terminal according to the determined scene mode.