Voice Detection by Multiple Networked Microphone Devices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In multi-device environments, conventional networked microphone devices (NMDs) often react independently to voice inputs, leading to inefficiencies in voice recognition and user experience, as they lack effective coordination to identify and process voice commands accurately across multiple devices.

Innovation Solution

A computing device identifies and processes a subset of voice recordings from multiple NMDs based on threshold sound pressure levels or zone configurations, refining and combining these recordings to determine a single voice command, thereby improving speech-to-text conversion and reducing redundant processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multiple NMDs process voice inputs independently, then each device can detect voice commands, but voice recognition accuracy and efficiency deteriorate due to redundant processing and lack of coordination

Engineering Contradiction:
Improvevoice recognition accuracyVSAvoidprocessing efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent combines voice recordings from multiple NMDs into a unified set of recordings for processing. Instead of each device operating independently, the system merges the audio data from all devices to create a consolidated voice command determination process, improving both accuracy and efficiency through coordinated processing.

Inventive Principle:
Principle #5Merging (Combining)

2Reliability

If all voice recordings from multiple NMDs are processed, then comprehensive voice detection is achieved, but processing time and computational resources increase

Engineering Contradiction:
Improvevoice command detection completenessVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts and processes only the necessary voice recordings from the set of recordings received from multiple NMDs. By identifying and processing only the relevant recordings needed to determine the voice command, the system avoids unnecessary processing of duplicate or redundant audio data, reducing processing time while maintaining detection completeness.

Inventive Principle:
Principle #2Taking out (Extraction)

3Adaptability or versatility

If multiple NMDs continuously monitor for voice inputs, then voice commands can be detected from any location, but energy consumption and processing overhead increase

Engineering Contradiction:
Improvespatial coverage for voice detectionVSAvoidenergy consumption per device
Core Design Contradiction:
Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent merges the monitoring capabilities of multiple NMDs into a coordinated system where devices share voice recording data through a network. This allows the system to maintain comprehensive spatial coverage while reducing individual device energy consumption, as each device can rely on others to detect and transmit voice inputs rather than all devices continuously processing audio locally.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP4036912B1Voice detection by multiple devices
Publication Date: 2024.12.04 SONOS INC
  • EP4036912B1 patent drawingFigure 1
  • EP4036912B1 patent drawingFigure 2~3
  • EP4036912B1 patent drawingFigure 4

AI summary

Disclosed herein are example techniques for voice detection by multiple NMDs. An example implementation may involve receiving a set of voice recordings from a set of NMDs, and identifying a subset of voice recordings from which to determine a given voice command. The example implementation may further involve causing the identified subset of voice recordings to be analyzed to determine the given voice command.