Temporal Noise Shaping

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

This invention introduces an improved method and apparatus for Temporal Noise Shaping (TNS) in audio encoding and decoding. The system intelligently selects between two types of filters—a high-energy filter and a low-energy (attenuated) filter—for each frame of an audio signal based on certain signal metrics (such as prediction gain). By adjusting the energy of the filter rather than simply turning TNS on or off, the system can maintain TNS’s advantages for audio quality while minimizing unwanted artifacts, especially at low bitrates or in frames prone to clicks/noise.

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

• Audio encoding and compression in music streaming services to improve sound quality at varying bitrates.
• Speech coding for telecommunication systems, including mobile voice calls and VoIP.
• Broadcast and digital radio systems that require high efficiency and audio fidelity.
• Hearing aid and assistive listening device signal processing.
• Audio codecs used in consumer electronics such as smartphones, Bluetooth speakers, and media players.

BenefitsContent extracted from patent full text and abstract with AI.

• Reduces unwanted audio artifacts (such as clicks and noise) caused by conventional TNS, especially at low bitrates or with specific audio signals.
• Maintains or improves perceived audio quality by dynamically choosing the optimal filter for each frame.
• Provides finer control over TNS processing, enabling better adaptation to signal characteristics.
• Enhances performance of audio codecs in challenging encoding scenarios (e.g., speech, tonal music) without sacrificing the benefits of TNS.
• Can be implemented in both hardware and software, and is compatible with existing audio coding standards.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

There are discussed methods and apparatus for performing temporal noise shaping. An apparatus may comprise a temporal noise shaping, TNS, tool (11) for performing linear prediction, LP, filtering (S33, S35, S36) on an information signal including a plurality of frames; and a controller (12) configured to control the TNS tool (11) so that the TNS tool (11) performs LP filtering with: a first filter (14a) whose impulse response has a higher energy (S36); and a second filter (15a) whose impulse response has a lower energy (S35), wherein the second filter is not an identity filter, wherein the controller (12) is configured to choose (S34) between filtering (S36) with the first filter (14a), and filtering (S35) with the second filter (15a) on the basis of a frame metrics.