Audio Time Stretch Apparatus for Jitter Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Internet real-time audio/video transmission, irregular packet time sequences due to jitter and clock drift lead to suboptimal digital-to-analog conversion, resulting in noticeable audio imperfections during playback.
Innovation Solution
An audio time stretch method that calculates energy levels of audio data, selectively performs waveform searches based on these levels, and adjusts audio data by removing or inserting data based on waveform similarities to maintain a stable digital-to-analog conversion sequence, using an apparatus with energy level, waveform search, determining, threshold, flag register, and buffer control modules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If audio time stretch is performed to correct irregular packet arrival times, then digital-to-analog conversion timing is improved, but audio quality degradation occurs due to noticeable imperfections
Solution Approach 1:
The patent applies local quality by performing waveform search and audio data adjustment only in low energy periods where imperfections are less noticeable. The system identifies specific time regions (low energy periods) and applies different processing strategies there versus high energy periods, making the time stretch operation local rather than global to minimize audible degradation.
Solution Approach 2:
The patent converts the harmful effect of low energy periods (where audio imperfections are more noticeable) into a benefit by selectively performing waveform search operations during these periods. The system uses the low energy characteristic to its advantage, applying time stretch adjustments when they are least likely to be perceived by listeners, thus turning a potential weakness into a strategic opportunity.
2Manufacturing precision
If waveform search is performed on all audio data, then time stretch accuracy is improved, but processing complexity and time consumption increase
Solution Approach 1:
The patent reduces processing complexity by applying waveform search selectively only to low energy periods rather than processing all audio data uniformly. This local approach maintains time stretch accuracy in critical regions while avoiding unnecessary computation in high energy periods where imperfections would be more noticeable anyway, thus resolving the contradiction between accuracy and complexity.
Solution Approach 2:
The patent applies partial action by performing waveform search on only a subset of audio data (specifically, low energy periods) rather than exhaustive processing of all data. This partial processing approach achieves sufficient time stretch accuracy for perceptual quality while significantly reducing computational burden and processing time.
3Reliability
If audio data is removed or inserted frequently to maintain timing, then packet timing synchronization is improved, but audio quality degradation increases due to excessive manipulation
Solution Approach 1:
The patent minimizes audio quality degradation by performing audio data removal and insertion operations locally only during low energy periods. By concentrating time stretch operations in these specific time regions rather than distributing them throughout the entire audio signal, the system maintains timing synchronization while reducing the total amount of manipulation that would otherwise be audible.
Solution Approach 2:
The patent converts the potential harm of frequent audio data manipulation into a benefit by strategically performing these operations during low energy periods. The low energy characteristic masks the effects of data removal and insertion, allowing the system to achieve better timing synchronization without proportionally increasing audible quality degradation.
Data Source
AI summary
An audio time stretch method and associated apparatus is provided. The method includes steps of calculating an energy level according to amplitudes of a plurality of received data, and determining whether the audio data requires audio time stretch according to the energy level. Audio data with lower energy level and volume are selectively time-stretched to alleviate audio quality degradation.