Head-Mounted Microphone Array Conversation Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional speech signal processing apparatuses face challenges in accurately determining whether a speaker in front is a conversing person due to the difficulty in separating the wearer's speech from the conversing person's speech when both are radiated in the same direction, affecting the accuracy of speech presence/absence and conversation establishment determination.
Innovation Solution
A conversation detection apparatus using a head-mounted microphone array that includes self-speech detection, front speech detection, and side speech detection sections, along with side direction conversation establishment degree calculation, to determine the presence of conversation in the front direction without relying on conversation establishment degree calculations affected by the wearer's speech, and controls directivity to suppress non-conversing person sound sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional speech signal processing apparatuses use microphone arrays to detect conversation, then they can identify sound sources, but they cannot accurately distinguish the wearer's speech from the conversing person's speech when both are radiated in the same direction
Solution Approach 1:
The patent segments the speech detection process into three independent detection sections: self-speech detection section for detecting the wearer's own speech, front speech detection section for detecting speech from the front direction, and side speech detection section for detecting speech from side directions. This segmentation allows each section to focus on specific speech sources and directions, enabling accurate differentiation between the wearer's speech and the conversing person's speech even when both are radiated in the same direction.
2Productivity
If the apparatus calculates conversation establishment degree in the front direction, then it can determine conversation presence, but the calculation is affected by the wearer's speech causing inaccurate results
Solution Approach 1:
The patent extracts and removes the wearer's speech component from the front direction speech signal before calculating the conversation establishment degree. The self-speech detection section identifies the wearer's speech, and this information is used to exclude the wearer's speech contribution from the front speech signal. This extraction ensures that the conversation establishment degree calculation in the front direction reflects only the conversing person's speech, eliminating the accuracy degradation caused by the wearer's speech being included in the calculation.
3Device complexity
If the apparatus uses only arriving direction determination to identify conversing persons, then it can process speech signals, but it cannot distinguish between words spoken to the wearer and words spoken to others in multi-conversation environments
Solution Approach 1:
The patent adds the dimension of speech source identification beyond just arriving direction by implementing separate self-speech detection and side speech detection sections. Instead of relying solely on the one-dimensional arriving direction information, the system incorporates additional detection dimensions: the wearer's own speech detection and side direction speech detection. This multi-dimensional approach enables the system to accurately identify conversing persons and distinguish between different conversation groups in complex acoustic environments with multiple simultaneous conversations.
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
Enables accurate detection of conversations in the front direction without being affected by the wearer's speech, improving the accuracy of determining the conversing person and enhancing speech detection and conversation establishment assessment.
Implementation Method 1
input signals from a microphone array
Implementation Method 2
form a directivity of sensitivity from input signals given by a plurality of microphone units
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A conversation detection apparatus uses a head-mounted microphone array to accurately determine whether a speaker in front is a conversing person or not. A conversation detection apparatus (100) includes a self-speech detection section (102) that detects a speech of a wearer of a microphone array (101), a front speech detection section (103) that detects a speech of a speaker in front of the microphone array wearer as a speech in front direction, a side speech detection section (104) that detects a speech of a speaker residing at at least one of right and left of the wearer as a side speech, a side direction conversation establishment degree deriving section (105) that calculates a conversation establishment degree between the speech of the wearer and the side speech, based on detection results of the speech of the wearer and the side speech, a front direction conversation detection section (106) that determines presence/absence of conversation in front direction based on a detection result of the front speech and a calculation result of the side direction conversation establishment degree, and an output sound control section (107) that controls directivity of speech heard by the hearing aid wearer, based on the determined presence/absence of conversation in front direction.