Adjustable Beamformer for Voice-Controlled Audio Channel Separation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing voice recognition-based control systems for audio recording devices lack real-time adjustable recording parameters and effective user control, particularly when capturing video scenes, as they often rely on cumbersome touch controls and fail to separate user commands from the main audio signal.

Innovation Solution

The implementation of an adjustable beamformer that defines virtual microphones for both audio recording and control channels, allowing voice commands to control audio attributes such as sensitivity, direction, and noise reduction, using a microphone array and beamforming techniques to isolate and process user commands separately from the main audio signal.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a directional microphone is used to capture user commands as an audio control channel, then voice command recognition is enabled, but the user commands are mixed with the main audio signal from the video scene

Engineering Contradiction:
Improvevoice command controlVSAvoidseparation of control signal from audio signal
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The patent divides the audio capture function into separate channels: a first audio channel captured by a first microphone for user commands, and a second audio channel captured by a second microphone for the video scene. This segmentation allows independent processing and preservation of both control signals and audio content without mutual interference.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts the user command signal from the mixed audio by using a dedicated first microphone specifically positioned and configured to capture only user commands. This extracted control signal is then processed separately through voice recognition while the second microphone captures the scene audio independently, preventing the control signal from contaminating the recorded audio.

Inventive Principle:
Principle #2Taking out (Extraction)

2Ease of operation

If touch control mechanisms are used for audio recording control, then user control capability is provided, but the control becomes cumbersome when trying to simultaneously capture a video scene

Engineering Contradiction:
Improveaudio recording controlVSAvoidtime to control audio recording
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The patent replaces mechanical touch control mechanisms with voice-based control. Users can issue audio attribute control commands through speech, which are recognized and processed by the system. This substitution eliminates the need for manual touch operations, allowing users to control audio recording parameters hands-free while simultaneously capturing video scenes.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If existing portable audio recorders are used, then basic audio recording is enabled, but real-time adjustable recording parameters are not provided to the user

Engineering Contradiction:
Improvereal-time adjustable recording parametersVSAvoidcontrol system complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a voice recognition system that can control multiple audio attributes including sensitivity, direction, and noise reduction through a single unified interface. This multi-functional control system allows users to adjust various recording parameters in real-time using voice commands, making the device adaptable to different recording scenarios without requiring separate control mechanisms for each parameter.

Inventive Principle:
Principle #6Universality (Multi-functionality)

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 real-time, user-definable audio recording attributes through voice control, improving user interaction by allowing precise control of audio recording parameters and reducing background noise, while maintaining clear separation of user commands from the recorded audio.

Implementation Method 1

an adjustable audio beamformer that defines a virtual microphone for at least one audio recording channel and for at least one audio control channel

Methodology Applied
Scientific EffectBeamforming:

Data Source

PatentEP3005349B1Voice controlled audio recording or transmission apparatus with adjustable audio channels
Publication Date: 2020.04.15 GOOGLE TECHNOLOGY HOLDINGS LLC
  • EP3005349B1 patent drawingFigure 1A
  • EP3005349B1 patent drawingFigure 1B
  • EP3005349B1 patent drawingFigure 2

AI summary

A method of operation beamforms a plurality of microphone outputs (112) to obtain a plurality of virtual microphone audio channels with at least one audio output channel and at least one audio control channel. The method performs voice recognition on the audio control channel to detect voice commands for controlling audio output channel attributes, and adjusts an audio channel attribute in response to detecting a voice command. Adjusting an attribute of the audio channel may be accomplished by, for example, controlling one or more parameters of an adjustable beamformer (120). The detected voice commands for controlling audio channel attributes may include voice commands for controlling audio sensitivity zooming, panning in a specified direction, focusing on a specified direction, blocking a specified direction, mixing a narrator's voice, blocking a narrator's voice, or reducing background noise. An apparatus that performs the method of operation is also disclosed.