Sound Discrimination via Harmonic Amplitude Differential
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sound discriminating devices that use audio to detect anomalies, such as screams, often produce false alarms due to misidentification of loud vowel sounds or distance-related volume issues, failing to correctly distinguish between intended and unintended sounds.
Innovation Solution
A sound discriminating device that extracts the differential value between the amplitude of an L-th harmonic and the amplitude of an (L+1)-th harmonic of the fundamental frequency, using this feature amount to determine if an input sound is a specific type of sound like a cry or anger, employing a computer program with a sound input unit, feature amount extracting unit, and determining unit, and utilizing acoustic models to accurately differentiate sounds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If volume-based scream detection is used, then detection sensitivity is improved, but false alarm rate increases due to misidentification of loud vowel sounds
Solution Approach 1:
The patent changes the detection parameter from volume alone to a combination of volume, duration, and fundamental frequency. This multi-parameter approach allows the system to distinguish between genuine screams and loud vowel sounds by checking if the fundamental frequency falls within the human voice range (85-255 Hz for males, 165-255 Hz for females), thereby reducing false alarms while maintaining detection sensitivity
Solution Approach 2:
The patent introduces fundamental frequency analysis as an intermediary criterion between volume detection and final scream identification. This intermediate step filters out false positives by verifying that the sound originates from human vocal cords, serving as a mediator that validates the initial volume-based detection
2Productivity
If fixed duration threshold is used for scream detection, then processing speed is improved, but detection accuracy deteriorates due to varied scream lengths
Solution Approach 1:
The patent applies dynamic threshold adjustment for duration based on the fundamental frequency detection results. When a sound meets the volume and frequency criteria, the duration threshold becomes more flexible, allowing the system to adapt to varying scream lengths while maintaining processing efficiency. This dynamic approach replaces fixed thresholds with condition-based thresholds
3Area of stationary object
If distance-independent detection is implemented, then coverage area is improved, but volume threshold setting becomes more difficult
Solution Approach 1:
The patent shifts from using absolute volume thresholds to using fundamental frequency as the primary discrimination parameter. Since fundamental frequency is inherent to the sound source and independent of distance, this parameter change enables distance-independent detection without requiring complex adaptive volume thresholding. The system detects screams across varying distances by relying on the stable frequency characteristic rather than distance-sensitive volume
Data Source
AI summary
A sound discriminating device capable of correctly discriminating a cry or other given sounds is provided. When a sound is input from a sound input unit, a feature amount extracting unit extracts a differential value between the amplitude of a fundamental frequency of the input sound and the amplitude of the second harmonic of the fundamental frequency as a feature amount of the input sound. A likelihood calculating unit calculates a likelihood between an acoustic model set for which a feature amount is known and the extracted feature amount. A result output unit determines whether or not the input sound is the given sound based on the result of the likelihood calculation.