Speech Intelligibility Measurement Using Multi-Microphone STI Averaging

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Large-scale networked audio systems in environments like conference rooms face challenges in tuning due to complexity, requiring expert teams and advanced test signal strategies to accurately configure and optimize speaker and microphone settings across multiple locations.

Innovation Solution

A method and apparatus that automatically identify and tune multiple speakers and microphones on a network by providing sequential test signals, detecting operational channels, establishing background noise levels, and applying infinite impulse response filters to equalize frequency responses, enabling automated setup and optimization without the need for expert teams.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If expert teams manually setup and test audio equipment, then configuration accuracy is improved, but installation time and cost increase

Engineering Contradiction:

Improveconfiguration accuracyVSAvoidinstallation time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The audio system performs self-configuration and self-testing through automated processes. The controller automatically identifies speakers and microphones, plays test signals, analyzes microphone responses, and adjusts audio parameters without requiring expert manual intervention, thereby reducing installation time while maintaining configuration accuracy

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual expert operations with automated electronic systems. The controller uses digital signal processing to generate test signals, capture microphone responses, and compute optimal audio settings, substituting the mechanical process of manual setup with an automated electronic measurement and adjustment system

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Manufacturing precision

If multiple speakers are tested sequentially with different test signals, then tuning accuracy is improved, but testing complexity increases

Engineering Contradiction:

Improvetuning accuracyVSAvoidtesting complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The testing process is segmented by assigning unique frequency components to different speakers. Each speaker receives a distinct test signal frequency, allowing the controller to individually identify and analyze the response of each speaker-microphone pair through frequency-based separation, simplifying the overall testing process while maintaining accuracy

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system uses periodic test signals with specific frequencies for each speaker. By employing sinusoidal test signals at different frequencies, the controller can systematically excite each speaker and measure the periodic response captured by microphones, enabling automated tuning through rhythmic, repeatable measurement cycles

Inventive Principle:

Principle #19Periodic action

3Productivity

If automated tuning is implemented, then setup time is reduced, but speech intelligibility measurement accuracy may worsen

Engineering Contradiction:

Improvesetup speedVSAvoidspeech intelligibility measurement accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system implements feedback by capturing microphone responses to test signals and using this information to automatically adjust audio parameters. The controller continuously measures the actual acoustic environment through microphone feedback and refines speaker settings based on the measured speech transmission index, ensuring accurate speech intelligibility assessment while maintaining automated operation

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The automated tuning process dynamically changes audio parameters such as speaker volume levels and equalization settings based on measured acoustic conditions. The system adjusts these parameters iteratively to optimize speech intelligibility, with the speech transmission index serving as the measurement criterion that guides parameter optimization throughout the automated setup process

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS11671065B2Measuring speech intelligibility of an audio environment

Publication Date: 2023.06.06 BIAMP SYST LLC

AI summary

An example method of operation may include initiating an automated tuning procedure, detecting via one or more microphones a sound measurement associated with an output of one or more speakers at two or more locations, determining a number of speech transmission index (STI) values equal to a number of microphones, and averaging the speech transmission index values to identify a single speech transmission index value.