Method and Apparatus for Spectrotemporally Improved Spectral Gap Filling in Audio Coding Using a Tilt

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

This patent describes methods and apparatus for improving how lost or missing parts of the audio spectrum (spectral gaps) are filled during audio coding and decoding. The core invention introduces a novel approach that uses adaptive 'spectral tilt' information, derived during encoding, to more precisely reconstruct the spectral envelope and energy distribution of the original audio signal when filling these gaps at the decoding stage. This approach compensates for the shortcomings of conventional techniques, especially at low bitrates, and results in audio output that more closely matches the original in both spectral and temporal characteristics.

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

• Audio and music streaming services aiming to deliver higher-quality sound at lower bitrates without noticeable artifacts or spectral holes.
• Speech codecs for telecommunication systems, especially where bandwidth is limited but speech intelligibility and audio quality must be preserved.
• Broadcasting (radio, TV) using advanced audio codecs for more efficient spectrum usage while maintaining sound quality.
• Hearing aids and assistive listening devices that perform real-time audio processing and require high-quality perceptual audio reconstruction at low bitrates.
• Mobile devices, such as smartphones and tablets, where storage and transmission efficiency for audio is essential.
• Voice-over-IP (VoIP) and video conferencing systems requiring efficient audio compression with natural and clear reproduction.

BenefitsContent extracted from patent full text and abstract with AI.

• Improves the perceived audio quality, especially in low bitrate scenarios, by more accurately reconstructing lost spectral content.
• Reduces signaling overhead compared to previous spectral gap filling techniques, making coding more efficient.
• Provides greater flexibility and adaptation by using frame-wise and signal-adaptive spectral tilt information instead of static parameters.
• Enhances both spectral (frequency) and temporal (time) fine structure reproduction, leading to more natural-sounding audio.
• Compatible with existing audio coding standards and can be implemented in both hardware and software, including as an upgrade to current codecs.
• Offers low-complexity computation, making it suitable for real-time applications and resource-constrained environments.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

Embodiments according to the invention are related to methods and apparatuses for spectrotemporally improved spectral gap filling in audio coding using a filtering. Embodiments according to the invention are related to methods and apparatuses for spectrotemporally improved spectral gap filling in audio coding using different noise filling methods. Embodiments according to the invention are related to methods and apparatuses for spectrotemporally improved spectral gap filling in audio coding using a tilt.