Vocal Stream Isolation via Destructive Interference and ICA
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions, such as conventional hearing aids and spatially directional systems, are inadequate in isolating specific vocal streams from acoustically noisy environments, particularly in settings where ambient noise interferes with speech comprehension, and are limited by size and effectiveness in environments with noise from similar frequency ranges.
Innovation Solution
A system that dynamically isolates and enhances target vocal signals by employing a tracking filter subsystem to selectively extract and suppress unwanted ambient noise, using principles of destructive interference and comb filtering to create a modified noise canceling signal that acoustically or electronically cancels ambient noise, allowing for clear discernment of targeted vocal streams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If conventional hearing aids amplify acoustic signals to compensate for hearing deficiencies, then the volume of speech is improved, but environmental noise is also amplified making discrimination difficult
Solution Approach 1:
The system segments the acoustic environment into multiple independent vocal streams using independent component analysis (ICA), separating desired speech from unwanted noise and competing conversations. This allows selective enhancement of specific speakers while suppressing others, resolving the contradiction between amplifying speech and avoiding noise amplification.
Solution Approach 2:
The system applies different processing characteristics to different spatial locations and frequency components. By creating location-specific transfer functions and applying spatial filtering, the system enhances speech from specific directions while attenuating noise from other directions, achieving local quality enhancement rather than uniform amplification.
2Ease of operation
If angular directionality is incorporated into hearing-assistance appliances to amplify signals from particular directions, then speech from specific directions is improved, but device size increases due to required large microphone arrays
Solution Approach 1:
The system replaces the mechanical approach of using large physical microphone arrays for spatial filtering with an electronic signal processing approach. By using ICA and adaptive filtering algorithms, the system achieves spatial discrimination and noise suppression computationally, eliminating the need for large physical apertures and enabling compact wearable device design.
3Measurement precision
If large microphone arrays are used to achieve substantial directionality for isolating vocal streams, then speech discrimination is improved, but the device becomes too large to be comfortably worn
Solution Approach 1:
The patent substitutes mechanical spatial filtering (large physical arrays) with electronic signal processing (ICA and adaptive filters). This computational approach achieves equivalent or superior vocal stream separation without requiring large physical dimensions, enabling the device to be compact and wearable while maintaining high discrimination precision.
Solution Approach 2:
The system changes the operating parameters from physical dimensions (array size) to signal processing parameters (filter coefficients, adaptation rates). By adjusting digital signal processing parameters rather than physical hardware dimensions, the system achieves high precision vocal stream discrimination in a compact form factor suitable for wearable applications.
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
Effectively enhances the ability to discern targeted vocal streams in noisy environments by suppressing unwanted noise, improving speech comprehension in real-time, even in settings with multiple overlapping vocal streams and ambient noise of similar frequency ranges, and can be implemented in small, wearable devices.
Implementation Method 1
employing a tracking filter subsystem to selectively extract and suppress unwanted ambient noise, using principles of destructive interference and comb filtering to create a modified noise canceling signal that acoustically or electronically cancels ambient noise
Implementation Method 2
employing a tracking filter subsystem to selectively extract and suppress unwanted ambient noise, using principles of destructive interference and comb filtering to create a modified noise canceling signal
Data Source
Figure 1~2
Figure 3~4
Figure 5~7
AI summary
The invention discloses systems and methods for enhancing the sound of vocal utterances of interest in an acoustically cluttered environment. The system generates canceling signals (sound suppression signals) for an ambient audio environment and identifies and characterizes desired vocal signals and hence a vocal stream or multiple streams of interest. Each canceling signal, or collectively, the noise canceling stream, is processed so that signals associated with the desired audio stream or streams are dynamically removed from the canceling stream. This modified noise canceling stream is combined (electronically or acoustically) with the ambient to effectuate a destructive interference of all ambient sound except for the removed audio streams, thus "enhancing" the vocal streams with respect to the unwanted ambient sound. Cepstral analysis may be used to identify a fundamental frequency associated with a voiced human utterance. Filtering derived from that analysis removes the voiced utterance from the canceling signal.