Early Reflections Filter Using Linear Prediction Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic devices face performance issues due to acoustically reflective surfaces, which affect speech recognition and sound quality by confusing sound source localization, as reflections from these surfaces interfere with direct audio signals.
Innovation Solution
The implementation of an Early Reflections Filter (ERF) using Multi-Channel Linear Prediction Coding (LPC) coefficients to suppress early reflections, allowing for better audio processing and improved sound source localization by attenuating acoustic reflections without altering the line-of-sight component.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If early reflections are suppressed using ERF with LPC coefficients, then sound source localization precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the audio signal processing into distinct components: capturing raw audio signals, generating multi-channel LPC coefficients, generating single-channel LPC coefficients, and combining them to create ERF filters. This segmentation allows each component to be optimized independently while working together to suppress early reflections and improve sound source localization precision.
Solution Approach 2:
The patent performs preliminary generation of multi-channel LPC coefficients from captured audio signals before the actual sound source localization processing. This preliminary action prepares the filtering components in advance, allowing the ERF to effectively suppress early reflections when needed, thereby improving localization precision without adding complexity to the critical localization path.
2Reliability
If ERF filtering is applied to suppress early reflections, then speech recognition performance is improved, but processing time increases
Solution Approach 1:
The patent changes the parameters of the audio signal processing by introducing ERF filters with specific frequency responses derived from LPC coefficients. These filters are designed to attenuate early reflections in specific frequency ranges while preserving speech frequencies, thereby improving speech recognition performance with minimal additional processing time.
Solution Approach 2:
The patent uses LPC (Linear Prediction Coding) to create simplified models of the audio signal characteristics. By copying and analyzing the spectral properties through LPC coefficients, the system can design ERF filters that effectively suppress reflections without requiring complex real-time processing, thus maintaining efficient processing time.
3Manufacturing precision
If multi-channel LPC coefficients are used to generate ERF filters, then audio processing quality is improved, but computational complexity increases
Solution Approach 1:
The patent merges multi-channel LPC coefficients with single-channel LPC coefficients to generate the ERF filters. This combining approach allows the system to leverage the spatial information from multi-channel processing while using single-channel processing for efficiency, achieving high audio processing quality without excessive computational complexity.
Solution Approach 2:
The patent applies ERF filtering selectively to suppress early reflections in specific frequency ranges and time windows, rather than processing the entire audio signal uniformly. This partial action approach focuses computational resources on the critical reflection suppression task, improving audio processing quality while managing computational complexity efficiently.
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
Enhances sound source localization and overall device performance by effectively filtering out early reflections, improving both speech recognition and sound quality in environments with acoustically reflective surfaces.
Implementation Method 1
the device uses an Early Reflections Filter (ERF) that makes use of Linear Prediction Coding (LPC), which is already being performed during speech processing, to suppress early reflections
Implementation Method 2
The device then uses this filter to generate filtered audio signals that suppress the early reflections
Data Source
AI summary
A system that performs early reflections filtering to suppress early reflections and improve sound source localization (SSL). During music playback and/or when a device is placed in a corner, acoustic reflections from nearby surfaces get boosted due to constructive interference, negatively impacting SSL and other processing of the device. To suppress these early reflections, the device uses an Early Reflections Filter (ERF) that makes use of Linear Prediction Coding (LPC), which is already being performed during speech processing. For example, the device generates raw audio signals using multi-channel LPC coefficients and then uses single-channel LPC coefficients for each raw audio signal in order to generate a filter that estimates the reflections. The device then uses this filter to suppress the early reflections and generate filtered audio signals, thus resulting in better audio processing and better overall device performance.