Generation of a downmix signal

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

This patent describes a device and method for downmixing audio signals (such as from stereo to mono or multichannel to fewer channels) in a way that minimizes undesirable audio artifacts. It achieves this by extracting and removing the parts of the second input signal that are highly similar (correlated) to the first input signal, and only adding the dissimilar (uncorrelated) parts to the first signal. This process reduces comb-filtering, signal loss, and other common problems when merging correlated audio signals.

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

• Mixing stereo or multichannel audio into mono or fewer channels for various audio reproduction systems.
• Audio mastering and production, where multiple tracks need to be combined cleanly.
• Audio compression codecs that require downmixing for efficient transmission or storage.
• Broadcasting and streaming audio services that must deliver content in different channel formats.
• Hearing aids or assistive listening devices that mix multiple microphones' signals for clarity.
• Voice communication systems or conferencing technologies that integrate several input channels into a single output.

BenefitsContent extracted from patent full text and abstract with AI.

• Reduces comb-filtering and other signal distortion artifacts common in traditional downmixing methods, improving audio quality.
• Preserves the overall energy and perceived loudness of the original signals, resulting in a natural-sounding mix.
• Computationally efficient, making it suitable for real-time processing and embedded devices.
• Flexible application that can be adapted for both software and hardware implementations.
• Allows fine control over mixing ratios and frequency-dependent adjustments, giving sound engineers greater creative control.
• Scalable: the method can be cascaded for multiple input channels beyond just two.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

An audio signal processing device (1) for downmixing of a first input signal ( X 1 ) and a second input signal ( X 2 ) to a downmix signal ( X D ) comprising: a dissimilarity extractor (2) configured to receive the first input signal ( X 1 ) and the second input ( X 2 ) signal as well as to output an extracted signal ( U 2 ), which is lesser correlated with respect to the first input signal ( X 1 ) than the second input signal ( X 2 ) and a combiner (3) configured to combine the first input signal ( X 1 ) and the extracted signal ( U 2 ) in order to obtain the downmix signal ( X D ).