Earpiece Voice Activity Detection Amid Audio Echo and Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Portable communication devices struggle to differentiate between a user's voice and background noise or music when handling multiple media experiences simultaneously, leading to difficulties in voice operation control.
Innovation Solution
The method and device employ an Ambient Sound Microphone and an Ear Canal Microphone to capture and process ambient and internal sounds, suppress echoes, and determine voice activity levels to control voice operations, using algorithms like normalized least mean squares for echo suppression and voice detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the microphone captures all sounds in the environment, then the device can detect user voice, but it also captures background noise and music making voice discrimination difficult
Solution Approach 1:
The patent segments the audio signal processing into multiple components: echo cancellation module, noise suppression module, and voice activity detection module. Each module handles a specific aspect of the audio signal, allowing the system to separate and process different sound sources independently, thereby improving voice detection accuracy in noisy environments
Solution Approach 2:
The patent introduces an intermediary processing layer between the microphone and the voice detection algorithm. This layer includes echo cancellers and noise suppressors that act as mediators to clean up the raw audio signal before it reaches the voice activity detector, reducing the harmful effects of background noise and echo
2Adaptability or versatility
If the device supports multiple media functions simultaneously, then the user experience is enhanced, but the device becomes confused about which media experience receives primary attention
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously monitors the audio environment and adjusts its behavior accordingly. The voice activity detection system provides feedback about user speech states, which then triggers appropriate media control actions, allowing the device to adaptively manage multiple media functions based on real-time conditions
Solution Approach 2:
The system employs self-service through automatic voice activity detection and context-aware media control. The device autonomously determines when the user is speaking and automatically adjusts media playback without requiring manual intervention, enabling seamless multi-media operation while reducing the complexity of user interaction
3Object-affected harmful factors
If audio processing technologies use noise suppression, then background noise is reduced, but the device remains sound agnostic and cannot differentiate between sounds
Solution Approach 1:
The patent applies different processing qualities to different parts of the audio signal. The noise suppression is applied selectively to background noise components while preserving the spectral and temporal characteristics of the user's voice. This localized processing approach reduces background noise without sacrificing the ability to differentiate between different sound sources
Data Source
AI summary
An earpiece (100) and a method (300) personalized voice operable control can include capturing (302) an ambient sound from an Ambient Sound Microphone (111) to produce an electronic ambient signal (426), delivering (304) audio content (402) to an ear canal (131) by way of an Ear Canal Receiver (125) to produce an acoustic audio content (404) and capturing (306) in the ear canal an internal sound (402) from an Ear Canal Microphone (123) to produce an electronic internal signal (410). The electronic internal signal includes an echo of the acoustic audio content and a spoken voice generated by a wearer of the earpiece. The Method also includes detecting (312) the spoken voice in the electronic internal signal in the presence of the echo, and controlling (314) a voice operation of the earpiece when the spoken voice is detected.