Subband Audio Encoding for High-Resolution Bitrate Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio data encoding methods for high-resolution audio with large sampling frequency and depth result in high bitrate, leading to inefficient encoding and transmission.
Innovation Solution
Perform subband decomposition on audio data to obtain multiple subband data items, determine a desired post-encoding bitrate for each item, and encode them to achieve a total bitrate equal to the desired bitrate of the audio data, using a well-established encoder like Opus.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If audio data with large sampling frequency and sampling depth is encoded using conventional methods, then high audio quality is maintained, but the bitrate becomes very large resulting in low encoding efficiency
Solution Approach 1:
The patent divides the audio data into multiple frequency subbands using subband decomposition. Each subband is then encoded separately with tailored encoding parameters. This segmentation allows the system to process different frequency ranges independently, achieving both high audio quality through precise frequency-specific encoding and improved encoding efficiency by distributing the processing load across multiple subbands rather than processing the entire high-resolution audio stream as a single block
2Manufacturing precision
If the sampling frequency and sampling depth of audio data are increased, then audio quality is improved, but the bitrate of encoded audio data becomes very large
Solution Approach 1:
The patent applies local quality by assigning different encoding parameters and bitrate allocations to different frequency subbands. Each subband receives encoding treatment optimized for its specific frequency characteristics, allowing the system to maintain high audio quality where it matters most while using more compressive encoding for less critical frequency ranges. This localized approach to quality management enables high-fidelity reproduction without requiring uniformly high bitrate across all frequency bands
Solution Approach 2:
The patent transforms the encoding parameters dynamically based on the frequency subband being processed. By changing encoding parameters such as bitrate allocation, window size, and transform type according to the specific subband characteristics, the system achieves efficient compression for each frequency range. This parameter adaptation allows high sampling frequency and depth data to be encoded at lower overall bitrates while maintaining the quality necessary for accurate audio reproduction
Data Source
Figure 1~2a
Figure 2b
Figure 3~4
AI summary
Embodiments of the present disclosure provide an audio data processing method and apparatus, an electronic device and a storage medium, wherein the method comprises: acquiring audio data to be processed whose sampling frequency is greater than or equal to a preset frequency and whose sampling depth is greater than or equal to a preset depth; performing subband decomposition on the audio data to obtain a plurality of subband data items; determining a desired post-encoding bitrate of each subband data item; a sum of desired post-encoding bitrates of all subband data items being equal to a desired post-encoding bitrate of the audio data; a ratio of a pre-encoding bitrate of the audio data to the desired post-encoding bitrate of the audio data being greater than a preset value; encoding each subband data item according to the desired post-encoding bitrate of each subband data item; generating encoded audio data according to each encoded subband data item.