Processor for Generating a Prediction Spectrum Based on Long-Term Prediction And/or Harmonic Post-filtering

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

This patent describes an advanced audio signal processing system that improves the quality of coded audio, especially for sounds with strong harmonic content (such as music or speech with prominent vowels). The invention uses a combination of long-term prediction (LTP) and/or harmonic post-filtering to enhance encoded audio by accurately predicting and amplifying harmonic structures even when pitch and amplitude vary rapidly. The approach splits audio into sub-intervals depending on detected pitch and processes them individually, allowing for adaptive, high-fidelity reconstruction of harmonics in both encoding and decoding stages. This yields better audio quality with efficient data usage, particularly for codecs that use the Modified Discrete Cosine Transform (MDCT).

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

• Audio compression standards and codecs for music streaming services (e.g., Spotify, Apple Music)
• Voice over IP (VoIP) and telephony for clearer, more natural speech at lower bitrates
• Hearing aids and assistive listening devices requiring high-quality sound processing
• Broadcasting (radio, TV) where bandwidth is limited but audio quality is critical
• High-quality audio transmission in gaming, video conferencing, or multimedia applications
• Archival and storage of audio materials with better preservation of musical and speech harmonics

BenefitsContent extracted from patent full text and abstract with AI.

• Improves the perceived quality of harmonic signals (such as music or speech vowels) especially at low bitrates
• Accurately tracks and adapts to rapid pitch and amplitude changes, overcoming the limitations of previous coding techniques
• Reduces quantization noise and artifacts between harmonics, resulting in cleaner audio output
• Allows for more efficient encoding by focusing more bits on unpredictable or transient regions of audio
• Operates both in time and frequency domains, enabling flexible and robust signal processing
• Can be implemented on both encoding and decoding sides, and is compatible with existing MDCT-based codecs
• Facilitates smoother transitions across sub-intervals for natural sound reconstruction and fewer perceptible artifacts
• Enables real-time application thanks to delay-less, adaptive processing

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

A processor for processing an (encoded) audio signal, the processor comprising:an LTP buffer configured to receive samples derived from a frame of the encoded audio signal;an interval splitter configured to divide a time interval associated with a subsequent frame of the encoded audio signal into sub-intervals depending on the encoded pitch parameter;calculation means configured to derive sub-interval parameters from the encoded pitch parameter dependent on a position of the sub-intervals within the time interval associated with the subsequent frame of the encoded audio signal;a predictor configured for generating a prediction signal from the LTP buffer dependent on the sub-interval parameters; anda frequency domain transformer configured for generating a prediction spectrum (XP) based on the prediction signal;and/or the processor comprising:a splitter configured for splitting a time interval associated with a frame of the audio signal into a plurality of sub-intervals, each having a respective length, the respective length of the plurality of sub-intervals being dependent on a pitch lag value;a harmonic post-filter configured for filtering the plurality of sub-intervals, wherein the harmonic post-filter is based on a transfer function comprising a numerator and a denominator, where the numerator comprises a harmonicity value, and wherein the denominator comprises a pitch lag value and the harmonicity value and/or a gain value.