Frequency-Domain Audio Coding With Backward-Compatible Block Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern frequency-domain speech/audio coding systems face challenges in achieving satisfactory quality for certain audio signals like rain or applause at low bitrates, as they either suffer from time-smearing due to long block coding or inefficiency with short block coding, leading to spectral holes.
Innovation Solution
A frequency-domain audio codec that supports transform length switching, allowing for the use of multiple transform lengths within a frame, with the ability to switch between them based on signalization, enabling backward compatibility with existing codecs while improving coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If long block coding is used, then temporal resolution is improved, but time-smearing of coding error (pre-echo) occurs
Solution Approach 1:
The frame is divided into multiple sub-frames, each processed with a separate transform. This segmentation allows the system to achieve fine temporal resolution in regions with transients while maintaining good quality in stationary regions, avoiding the time-smearing problem of long blocks without requiring uniformly short blocks throughout.
2Object-generated harmful factors
If short block coding is used, then time-smearing is reduced, but coding efficiency decreases due to increased data overhead
Solution Approach 1:
Different transform lengths are applied to different sub-frames based on local signal characteristics. Sub-frames containing transients use short transforms to minimize pre-echo, while stationary sub-frames use long transforms to maintain coding efficiency. This local adaptation resolves the contradiction by applying short blocks only where necessary.
3Adaptability or versatility
If a new frequency-domain audio codec supporting multiple transform lengths is developed, then adaptability to different audio signals is improved, but device complexity and market adoption difficulty increase
Solution Approach 1:
The codec dynamically selects transform lengths based on signal characteristics detected in the audio frame. The system adapts the transform configuration frame-by-frame or sub-frame-by-sub-frame, allowing a single codec implementation to handle multiple signal types (transient, stationary, mixed) without requiring multiple fixed codec variants.
4Manufacturing precision
If transform length switching is implemented, then audio quality for transient signals is improved, but backward compatibility with existing codecs becomes challenging
Solution Approach 1:
The codec is designed with multi-functionality to handle both traditional long-block mode and the new transform-length-switching mode. The same decoder architecture can process bitstreams encoded with either approach, ensuring backward compatibility while enabling improved quality when the switching capability is utilized.
Data Source
AI summary
A frequency-domain audio codec is provided with the ability to additionally support a certain transform length in a backward-compatible manner, by the following: the frequency-domain coefficients of a respective frame are transmitted in an interleaved manner irrespective of the signalization signaling for the frames as to which transform length actually applies, and additionally the frequency-domain coefficient extraction and the scale factor extraction operate independent from the signalization. By this measure, old-fashioned frequency-domain audio coders/decoders, insensitive for the signalization, would be able to nevertheless operate without faults and with reproducing a reasonable quality. Concurrently, frequency-domain audio coders/decoders able to support the additional transform length would offer even better quality despite the backward compatibility. As far as coding efficiency penalties due to the coding of the frequency domain coefficients in a manner transparent for older decoders are concerned, same are of comparatively minor nature due to the interleaving.