Sound Signal Thresholding for Speech Detection Under Rising Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting a detection target in a sound signal, such as speech, are inaccurate when noise power abruptly rises, causing noise to be misidentified as the target.
Innovation Solution
An information processing device that segments a sound signal into sections, calculates variation values, sets a detection threshold based on the maximum power of identified sections, and detects sections where the power exceeds this threshold as the detection target.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a threshold value based on noise power is used for detection, then detection simplicity is maintained, but detection accuracy deteriorates when noise power rises abruptly
Solution Approach 1:
The patent applies preliminary action by segmenting the sound signal into multiple sections and calculating variation values before setting the detection threshold. This preprocessing step identifies sections with low variation (likely noise) and uses their maximum power to establish a threshold that anticipates noise fluctuations, thereby maintaining detection simplicity while improving accuracy against abrupt noise rises.
Solution Approach 2:
The patent segments the sound signal into multiple sections and calculates variation values for each segment. By dividing the signal processing into discrete sections with different characteristics (low variation for noise, high variation for speech), the system can set thresholds based on noise sections while maintaining overall detection simplicity, resolving the contradiction between ease of operation and measurement precision.
2Productivity
If a fixed threshold value is used for detection, then detection speed is maintained, but detection reliability deteriorates when noise conditions change
Solution Approach 1:
The patent implements dynamics by making the detection threshold adaptive rather than fixed. The threshold is dynamically determined based on the maximum power among sections identified as noise (low variation sections). This allows the threshold to automatically adjust to changing noise conditions while maintaining detection speed, as the adaptation is based on pre-calculated variation values and section powers.
Solution Approach 2:
The system performs self-service by automatically adapting the detection threshold based on the signal's own characteristics. The variation value calculation and section identification process enables the system to self-determine which sections represent noise conditions and use their maximum power to set an appropriate threshold, eliminating the need for external threshold configuration and improving reliability under varying noise conditions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information processing device (100) includes an acquisition unit (110) that acquires a sound signal and a control unit (120) that segments the sound signal into a plurality of sections, calculates a variation value as a variation amount per section time in regard to each of the plurality of sections based on the sound signal, identifies sections where the variation value is less than or equal to a predetermined threshold value among the plurality of sections, calculates power of the sound signal in each of the identified sections based on the sound signal, determines a maximum value among values of the power of the sound signal in each of the identified sections, sets a value based on the maximum value as a detection threshold value, and detects sections where the power of the sound signal with elapse of time is higher than or equal to the detection threshold value as detection target sections.