Multichannel Acoustical Signal Time Companding via Common Pitch Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for time companding of multichannel acoustical signals, such as stereo or 5.1 channel signals, face challenges in accurately extracting a common feature (pitch) across channels, leading to phase differences and discomfort due to out-of-phase signals when combining channels for processing.
Innovation Solution
An acoustical-signal processing apparatus and method that extracts feature data common to all channels by calculating composite similarities from each channel signal, using techniques like the PICOLA method for time compression and expansion, ensuring synchronization across channels through a common fundamental frequency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a feature common to all channels is extracted from signals combining (adding) multichannel acoustical signals, then synchronization between channels can be secured, but a feature (common pitch) cannot be accurately extracted when there is included a sound having a component of a left channel out of phase with that of a right channel
Solution Approach 1:
The patent divides the feature extraction process into separate operations for each channel. Instead of combining signals first and then extracting features, the system extracts feature data (such as fundamental frequency and timbre) separately from each channel signal, then uses these individual features to determine common pitch. This segmentation allows accurate feature extraction from out-of-phase signals while maintaining channel synchronization through the common pitch information.
2Adaptability or versatility
If time-base companding is separately executed for each channel based on individually extracted feature data, then processing flexibility is improved, but phase differences are caused between channels leading to discomfort
Solution Approach 1:
The patent introduces common pitch as an intermediary element that mediates between individual channel features and the final time-base companding operation. The system extracts feature data from each channel, determines common pitch from these features, and then uses the common pitch to control the companding process for all channels. This intermediary ensures that while each channel can be processed with its own characteristics, the overall phase coherence is maintained through the shared common pitch reference.
3Productivity
If conventional time companding methods are used with adaptive time width based on feature data, then time compression and expansion can be realized, but computational complexity increases and time resolution may be reduced
Solution Approach 1:
The patent changes the approach to time width determination by using fixed or predetermined time widths based on the extracted feature data (such as fundamental frequency periods) rather than complex adaptive time width calculations. The system determines appropriate time widths from the feature data itself (e.g., using the period of the fundamental frequency) and applies these to the overlap-add operation. This parameter change simplifies the computational process while maintaining accurate time compression and expansion effects.
Data Source
AI summary
An acoustical-signal processing apparatus includes a feature extracting unit that extracts feature data common to each channel signal which forms a multichannel acoustical signal, based on a composite similarity obtained by combining similarities calculated from each channel signal; and a time-base companding unit that executes time compression and time expansion of the multichannel acoustical signal based on the extracted feature data.