Headphone Vibration Sensing for Accurate Wearer Utterance Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for detecting utterances using acceleration sensors in headphones are prone to errors due to vibration interference from loudspeaker output, leading to false detections when human voice is present in music or noise.
Innovation Solution
A signal processing apparatus and method that includes noise generation and addition, vibration calculation, or transmission component prediction and subtraction to differentiate and mask vibration patterns, ensuring accurate detection of wearer utterances by using a vibration sensor.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If large volume sound is output from the loudspeaker, then the vibration reproduction performance is improved, but the utterance detection accuracy deteriorates due to vibration transmission to the acceleration sensor
Solution Approach 1:
The patent segments the vibration signal into two components: one caused by sound output from the loudspeaker and another caused by utterance. By separating these vibration sources and processing them independently, the system can maintain accurate utterance detection even when the loudspeaker is operating at high volume.
Solution Approach 2:
The patent introduces a vibration pattern generation unit that creates a reference vibration pattern based on the sound output signal. This reference pattern acts as an intermediary to compare against the actual acceleration sensor signal, enabling the system to distinguish between vibrations caused by sound output and those caused by utterance.
2Difficulty of detecting and measuring
If the vibration sensor detects both sound output vibration and utterance vibration, then the sensitivity of the detection system is improved, but the reliability of utterance detection deteriorates due to false positives
Solution Approach 1:
The patent uses feedback by continuously comparing the actual acceleration sensor signal with the generated reference vibration pattern. The comparison result feeds back to determine whether an utterance has occurred, allowing the system to maintain high sensitivity while filtering out false positives caused by sound output vibrations.
Solution Approach 2:
The patent converts the harmful effect of sound output vibration into a beneficial reference signal. By generating a vibration pattern from the sound output signal itself, the system transforms the interference into a useful comparison baseline that helps identify genuine utterances.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Effectively distinguishes between wearer-generated vibrations and interference from loudspeaker output, enhancing the accuracy of utterance detection in noisy environments.
Implementation Method 1
a vibration sensor 140 that detects vibration
Data Source
AI summary
Provided is a signal processing apparatus, signal processing method, and program capable of detecting an utterance by a wearer even in a state where sound is output from a vibration reproduction apparatus. A signal processing apparatus including a processing unit that operates corresponding to a vibration reproduction apparatus including a vibration reproduction unit that reproduces vibration and a vibration sensor that detects vibration, and performs processing of making it difficult to detect an utterance in utterance detection processing of detecting an utterance by a wearer of the vibration reproduction apparatus on the basis of the vibration sensor signal.