Method and Apparatus for Signal-Adaptive Transform Kernel Switching in Audio Coding

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

This patent introduces a method and apparatus for audio coding and decoding that adaptively selects between different transform kernels based on the characteristics of the audio signal. By switching between several types of lapped transforms (such as various forms of Modified Discrete Cosine and Sine Transforms with different symmetry properties), the system improves audio encoding efficiency, particularly in challenging cases such as highly harmonic signals or stereo signals with significant phase differences. The system transmits side information that indicates which transform kernel is used for each block or frame, ensuring accurate decoding and perfect reconstruction of the original audio.

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

• Advanced audio codecs for music, speech, and broadcasting applications
• Streaming services (like Spotify or Apple Music) that require efficient music compression
• Mobile devices and DSPs where computational and bandwidth efficiency are critical
• Professional audio and video editing software needing high-fidelity audio coding
• Telecommunications codecs for improved call quality and efficiency (e.g., VoIP)
• Audio coding in TV and radio broadcasting, especially with multi-channel and surround sound formats
• Storage of high-quality, compressed audio files for digital distribution

BenefitsContent extracted from patent full text and abstract with AI.

• Higher audio quality at the same or lower bitrates, especially for difficult-to-code signals
• Improved coding efficiency for harmonic signals and challenging stereo/multichannel content
• Reduced computational complexity by using signal-adaptive switching rather than complex auxiliary tools
• Flexible and robust handling of various audio types, bitrates, and channel configurations
• Enables more efficient use of bandwidth in streaming and broadcasting
• Maintains perfect reconstruction (no loss) despite switching between transform types
• Easily integrates into existing and emerging audio coding standards (such as MPEG-H 3D Audio)

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

A schematic block diagram of a decoder 2 for decoding an encoded audio signal 4 is shown. The decoder comprises an adaptive spectrum-time converter 6 and an overlap-add-processor 8. The adaptive spectrum-time converter converts successive blocks of spectral values 4' into successive blocks of time values 10, e.g. via a frequency-to-time transform. Furthermore, the adaptive spectrum-time converter 6 receives a control information 12 and switches, in response to the control information 12, between transform kernels of a first group of transform kernels comprising one or more transform kernels having different symmetries at sides of a kernel, and a second group of transform kernels comprising one or more transform kernels having the same symmetries at sides of a transform kernel. Moreover, the overlap-add-processor 8 overlaps and adds the successive blocks of time values 10 to obtain decoded audio values 14, which may be a decoded audio signal.