Apparatus and Method for Removing Undesired Auditory Roughness

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

This patent describes an apparatus, method, and encoder/decoder system for reducing undesired auditory roughness in audio signals, particularly those processed using low-bitrate codecs. The invention analyzes spectral components of an audio signal to detect roughness, which often arises due to quantization errors or spectral band replication in audio compression. It selectively attenuates or removes tonal side-peaks that cause perceptible roughness, guided by side information generated during encoding and transmitted in the audio bitstream, thereby preserving the perceived quality of the reconstructed audio output.

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

• Improving audio quality in music and speech codecs for streaming services operating at low bitrates (e.g., for mobile or bandwidth-limited environments).
• Post-processing of compressed audio in consumer electronic devices (smartphones, tablets, Bluetooth speakers) to reduce audio artifacts.
• Enhancing audio codecs in telecommunications and VoIP applications for clearer, high-fidelity calls at low bitrates.
• Audio mastering and production tools aiming to clean up artifacts in compressed musical content.
• Hearing aids or assistive listening devices that use real-time audio compression and need artifact mitigation.

BenefitsContent extracted from patent full text and abstract with AI.

• Significantly improves subjective audio quality by reducing perceptual roughness artifacts introduced during compression, especially at very low bitrates.
• Works on a wide range of audio content, including polyphonic (multi-pitch) signals, overcoming limitations of previous roughness removal methods that only supported mono-pitch signals.
• Requires minimal increase in bitrate thanks to efficient side-information signaling (e.g., ~0.3–0.6 kbps overhead).
• Operates selectively based on psychoacoustic modeling to avoid unnecessary signal modification, preserving naturalness of the original audio.
• Adaptable for integration into existing codec frameworks and audio devices, and implementation is possible in hardware, software, or hybrid forms.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

An apparatus (100) for processing an audio input signal to obtain an audio output signal according to an embodiment. The apparatus (100) comprises a signal analyser (110) configured for determining information on an auditory roughness of one or more spectral bands of the audio input signal. Moreover, the apparatus (100) comprises a signal processor (120) configured for processing the audio input signal depending on the information on the auditory roughness of the one or more spectral bands.