Encoder and Method for Encoding an Audio Signal with Reduced Background Noise Using Linear Predictive Coding

Simple SummaryContent extracted from patent full text and abstract with AI.

This patent describes an audio encoder and method that reduces background noise in audio signals using linear predictive coding (LPC). By estimating and subtracting background noise directly in the time domain, and using a combination of LPC filters, the encoder jointly enhances and compresses the audio signal. This approach avoids the delays and complexity of traditional frequency-domain noise reduction, allowing real-time high-quality speech encoding, especially useful for communication devices.

Use CasesContent extracted from patent full text and abstract with AI.

• Mobile phones and hands-free communication systems to improve speech clarity in noisy environments.
• Teleconferencing and VoIP (Voice over IP) platforms for real-time communication without perceptible delay.
• Speech codecs in automotive infotainment and in-car communication systems.
• Hearing aids or assistive listening devices that require low-latency, noise-reduced audio processing.
• Broadcast and live streaming where real-time speech encoding with background noise reduction is essential.
• Smart home devices and voice assistants for clearer voice recognition.

BenefitsContent extracted from patent full text and abstract with AI.

• Reduces background noise without introducing significant delay, enabling real-time or near real-time audio processing.
• Jointly optimizes noise reduction and audio encoding, minimizing both interference and quantization noise for better audio quality.
• Operates in the time domain, eliminating the need for complex and delay-inducing frequency domain transformations (like overlap-add windowing).
• Decreases computational complexity compared to traditional cascaded noise reduction and encoding approaches.
• Improves both objective and subjective speech quality (demonstrated through SNR improvements and listening tests).
• Suitable for implementation in software, hardware, or embedded systems, offering design flexibility for various applications.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

It is shown an encoder for encoding an audio signal with reduced background noise using linear predictive coding. The encoder comprises a background noise estimator configured to estimate background noise of the audio signal, a background noise reducer configured to generate background noise reduced audio signal by subtracting the estimated background noise of the audio signal from the audio signal, and a predictor configured to subject the audio signal to linear prediction analysis to obtain a first set of linear prediction filter (LPC) coefficients and to subject the background noise reduced audio signal to linear prediction analysis to obtain a second set of linear prediction filter (LPC) coefficients. Furthermore, the encoder comprises an analysis filter composed of a cascade of time-domain filters controlled by the obtained first set of LPC coefficients and the obtained second set of LPC coefficients.