Audio Encoder, Audio Decoder, Method for Encoding an Audio Signal and Method for Decoding an Encoded Audio Signal

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

This patent describes an improved method and system for encoding and decoding audio signals using predictive techniques that operate on spectral coefficients in the transform (frequency) domain. Rather than applying prediction to all frequency bins, the invention selectively predicts only the coefficients associated with harmonic (tonal) components, using a 'spacing value' that denotes harmonic structure (e.g., fundamental pitch and its multiples). This makes the process more efficient by focusing on predictable, tonal parts of the audio and leaving out noisy, less predictable regions.

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

• Audio compression for music streaming services, enabling higher quality at lower bitrates.
• Speech coding in telecommunications (such as mobile phones or VoIP) to improve clarity and reduce data use.
• Professional audio production and broadcasting, where efficient storage or transmission of high-quality audio is important.
• Encoding for 3D audio or immersive sound experiences in VR/AR environments.
• Implementation in audio formats and codecs such as MPEG-H 3D, Advanced Audio Coding (AAC), or next-gen audio standards.

BenefitsContent extracted from patent full text and abstract with AI.

• Reduces the computational complexity and processing power needed for audio coding/decoding, especially compared to methods predicting all spectral bins.
• Increases the efficiency of predictive encoding by targeting only tonal (harmonic) components, thereby providing higher audio quality at low bitrates.
• Minimizes unwanted prediction of non-tonal/noisy frequency regions, leading to fewer reconstruction errors and artifacts in the decoded audio.
• Decreases algorithmic delay, making it suitable for real-time, low-latency communications and applications.
• Simplifies synchronization between encoder and decoder, as the selection of predicted coefficients relies on transmitted spacing values rather than continuous adaptation.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

An encoder for encoding an audio signal. The encoder is configured to encode the audio signal in a transform domain or filter-bank domain, wherein the encoder is configured to determine spectral coefficients of the audio signal for a current frame and at least one previous frame, wherein the encoder is configured to selectively apply predictive encoding to a plurality of individual spectral coefficients or groups of spectral coefficients which are separated by at least one spectral coefficient.