Dual-Microphone Voice Control for Noisy Earpiece Speech Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Earpiece devices struggle to differentiate between user voice and background noise in noisy environments, leading to degraded communication quality due to their inability to discriminate between voices in the background and the user's voice.
Innovation Solution
A device with dual microphones and a processor that analyzes spectral characteristics to learn and predict the user's voice in noisy environments, allowing for voice-operated control by mixing and filtering signals to enhance the user's voice while reducing background noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If audio processing technologies are used to suppress noise, then noise suppression is achieved, but the earpiece cannot differentiate between user's voice and other people's voices in the background
Solution Approach 1:
The patent divides the audio processing task into multiple stages: initial noise suppression, followed by voice activity detection, and finally selective amplification of the user's voice. This segmentation allows the system to first reduce overall noise levels, then identify and preserve the user's voice components while suppressing other background voices.
Solution Approach 2:
The patent introduces voice activity detection as an intermediary process between noise suppression and audio output. This intermediary layer analyzes the suppressed audio signal to identify segments containing the user's voice, enabling selective enhancement without amplifying background noise or other people's voices.
2Quantity of substance
If the earpiece microphone captures all sounds in the environment, then comprehensive sound capture is achieved, but the quality of communication is degraded by background sounds
Solution Approach 1:
The patent implements dynamic gain adjustment based on voice activity detection. The system continuously monitors the captured audio signal and dynamically adjusts the gain for different time segments, amplifying portions containing the user's voice while attenuating portions dominated by background noise, thus maintaining both comprehensive capture and high communication quality.
Solution Approach 2:
The patent changes the audio signal parameters (gain, amplification level) based on the detected voice activity. By monitoring spectral characteristics and energy levels across different frequency bands, the system adjusts parameters in real-time to enhance the user's voice while suppressing background sounds, resolving the contradiction between comprehensive capture and communication quality.
3Device complexity
If the earpiece is sound agnostic and cannot differentiate sounds, then simple processing is achieved, but the user's voice cannot be distinguished from other voices in proximity
Solution Approach 1:
The patent performs preliminary voice activity detection and spectral analysis on the captured audio signal before final processing. By pre-identifying segments containing the user's voice and characterizing their spectral properties, the system prepares the data for selective enhancement, achieving accurate voice differentiation without requiring complex real-time processing during audio output.
Data Source
AI summary
At least one exemplary embodiment is directed to a method and device for voice operated control with learning. The method can include measuring a first sound received from a first microphone, measuring a second sound received from a second microphone, detecting a spoken voice based on an analysis of measurements taken at the first and second microphone, learning from the analysis when the user is speaking and a speaking level in noisy environments, training a decision unit from the learning to be robust to a detection of the spoken voice in the noisy environments, mixing the first sound and the second sound to produce a mixed signal, and controlling the production of the mixed signal based on the learning of one or more aspects of the spoken voice and ambient sounds in the noisy environments.