Speech Signal Processing for Intelligibility via Harmonic Gain
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sound quality enhancing algorithms face a tradeoff between minimizing residual background noise and speech distortion, which can deteriorate speech intelligibility.
Innovation Solution
A speech signal processing apparatus and method that determines a gain for input signals based on harmonic characteristics of voiced speech using a comb filter, preserves voiced speech by applying this gain, and uses linear predictive coefficients to preserve unvoiced speech, effectively reducing background noise while minimizing distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If background noise is removed from input signal, then residual background noise is reduced, but speech distortion is intensified and speech intelligibility deteriorates
Solution Approach 1:
The patent segments the speech signal into voiced speech and unvoiced speech components, applying different processing strategies to each. Voiced speech undergoes harmonic characteristic-based gain determination to preserve musicality, while unvoiced speech is processed separately using linear predictive coefficients. This segmentation allows selective noise reduction without uniformly distorting all speech components.
Solution Approach 2:
The patent applies local quality by determining gain values specifically for harmonic components of voiced speech based on their spectral characteristics. The comb filter is designed to target specific frequency regions where harmonic structures exist, allowing noise reduction in those regions while preserving the local quality and intelligibility of the voiced speech portions.
2Loss of information
If gain is applied to preserve harmonic components, then speech intelligibility is enhanced, but speech distortion may occur
Solution Approach 1:
The patent employs feedback mechanisms where the processed speech signal is continuously monitored and fed back into the system. The gain values are dynamically adjusted based on the feedback from harmonic characteristic analysis, allowing the system to learn from previous processing results and refine its performance to minimize distortion while maintaining intelligibility.
Solution Approach 2:
The patent changes parameters dynamically by adjusting gain values based on the detected harmonic characteristics of the input signal. The comb filter parameters are modified according to the spectral content, and linear predictive coefficients are updated in real-time. This parameter adaptation allows the system to optimize speech intelligibility while minimizing distortion for each specific input condition.
Data Source
AI summary
A speech signal processing apparatus and a speech signal processing method for enhancing speech intelligibility are provided. The speech signal processing apparatus includes an input signal gain determiner to determine a gain of an input signal based on a harmonic characteristic of a voiced speech, a voiced speech output unit to output a voiced speech in which a harmonic component is preserved by applying the gain to the input signal, a linear predictive coefficient determiner to determine a linear predictive coefficient based on the voiced speech, and an unvoiced speech preserver to preserve an unvoiced speech of the input signal based on the linear predictive coefficient.