Apparatus and Method for Generating an Encoded Signal or for Decoding an Encoded Audio Signal Using a Multi Overlap Portion

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

This patent describes a method and apparatus for encoding and decoding audio and image signals using a multi-overlap region, where at least three window functions overlap during the signal processing stage (such as during Modified Discrete Cosine Transform (MDCT)-based coding). The invention enables more flexible and adaptive signal windowing, particularly for handling transients (sudden changes in the signal), by dynamically adjusting the overlap and length of transform windows. It reduces processing delay, improves the quality around signal transients, and minimizes audible artifacts (such as pre-echo and post-echo) that are common in compressed audio and images. Both encoding and decoding processes are synchronized using window sequence information, ensuring efficient and artifact-free signal coding and reconstruction.

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

• Low-delay speech and audio codecs for real-time communications (e.g., voice or video calls).
• High-quality audio compression for streaming services (music, podcasts, live audio).
• Efficient video and image compression to reduce blocking artifacts, useful in video conferencing, broadcasting, and online streaming.
• Adaptive audio and image coding in mobile devices, enabling high quality with lower processing delays and reduced memory usage.
• Professional audio production where precise handling of transients is critical (e.g., recording studios or live sound engineering).
• Gaming and VR/AR, where synchronized, low-latency audio/video playback is required.

BenefitsContent extracted from patent full text and abstract with AI.

• Significantly reduces encoding/decoding delay (low-latency), important for real-time applications.
• Provides higher audio and image quality by adapting window overlaps and transform lengths based on transient detection.
• Minimizes compression artifacts such as pre-echo and post-echo, which improves perceived signal quality, especially during sharp attacks or transients.
• Greater flexibility compared to binary (on/off) window overlap schemes, allowing more granular adaptation to signal content.
• Synchronizes encoder and decoder with the same windowing schemes and window shapes, reducing implementation complexity and memory usage (especially valuable for mobile and embedded devices).
• Compatible with existing transform coding frameworks (e.g., MDCT), enabling easier integration into standard audio/image codecs.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

An apparatus for generating an encoded signal, comprises: a window sequence controller (808) for generating a window sequence information (809) for windowing an audio or image signal, the window sequence information indicating a first window (1500) for generating a first frame of spectral values, a second window function (1502) and at least one third window function (1503) for generating a second frame of spectral values, wherein the first window function (1500), the second window function (1502) and the one or more third window functions overlap within a multi-overlap region (1300); a preprocessor (802) for windowing (902) a second block of samples corresponding to the second window function and the at least one third window functions using an auxiliary window function (1 100) to obtain a second block of windowed samples, and for preprocessing (904) the second block of windowed samples using a folding-in operation of a portion of the second block overlapping with a first block into the multi-overlap portion (1300) to obtain a preprocessed second block of windowed samples having a modified multi-overlap portion; a spectrum converter (804) for applying an aliasing-introducing transform (906) to the first block of samples using the first window function to obtain the first frame of spectral values, for applying the aliasing introducing transform to a first portion of the preprocessed second block of windowed samples using the second window function to obtain a first portion of spectral samples of a second frame and for applying the aliasing introducing transform to a second portion of the preprocessed second block of windowed samples using the one or more third window functions (1503) to obtain a second portion of spectral samples of the second frame; and a processor (806) for processing the first frame and the second frame to obtain encoded frames of the audio or image signal.