Audio Transient Manipulation via Segmentation and Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio signal processing methods, such as phase vocoders and (P)SOLA, often distort transient events during time stretching or pitch manipulation, leading to blurred or unnatural sound quality due to the destruction of vertical coherence, resulting in temporal dispersion and artifacts like pre- and post-echoes.

Innovation Solution

The method involves removing transient portions before processing, either by copying or synthesizing them, and then reinserting them accurately after processing to maintain the original transient characteristics, ensuring that the manipulated signal preserves the subjective quality of the original transient events.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If phase vocoders or (P)SOLA methods are used for time stretching or pitch manipulation, then the reproduction speed or pitch is changed, but the transient quality is degraded due to destruction of vertical coherence

Engineering Contradiction:
Improvereproduction speedVSAvoidtransient quality
Core Design Contradiction:
SpeedVSManufacturing precision

Solution Approach 1:

The audio signal is segmented into transient portions and non-transient portions using a transient detector. The transient portions are extracted and set aside before processing, while only the non-transient portions are subjected to phase vocoder or (P)SOLA processing. After processing, the transient portions are reinserted into the processed signal at their original positions, preserving their vertical coherence while still allowing speed or pitch changes in the overall signal.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The transient portions are extracted from the audio signal using a transient detector that identifies regions with rapidly changing energy distribution across frequency bands. These extracted transient portions are temporarily removed from the processing chain, allowing the main signal processing to occur without affecting them. The transients are then reinserted after processing to maintain their original characteristics.

Inventive Principle:
Principle #2Taking out (Extraction)

2Adaptability or versatility

If phase vocoders or (P)SOLA methods are used for time stretching or pitch manipulation, then the reproduction speed or pitch is changed, but temporal dispersion and echo artifacts occur

Engineering Contradiction:
Improvepitch manipulation capabilityVSAvoidtemporal dispersion and echo artifacts
Core Design Contradiction:
Adaptability or versatilityVSObject-generated harmful factors

Solution Approach 1:

The signal is divided into transient and non-transient segments. The phase vocoder or (P)SOLA processing is applied only to non-transient segments, avoiding the generation of temporal dispersion and echo artifacts in transient regions. The transient segments are preserved separately and reinserted, eliminating the harmful artifacts while maintaining pitch manipulation capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Transient portions that would generate harmful artifacts when processed are extracted and removed from the processing path. This prevents the generation of temporal dispersion and echo artifacts in these critical regions, while the non-transient portions still receive the full pitch manipulation treatment.

Inventive Principle:
Principle #2Taking out (Extraction)

3Ease of operation

If the vertical coherence of transients is destroyed by frequency-dependent processing, then the signal is manipulated in stationary portions, but the transient portions have reduced quality

Engineering Contradiction:
Improvesignal processing capabilityVSAvoidtransient quality
Core Design Contradiction:
Ease of operationVSManufacturing precision

Solution Approach 1:

The audio signal is segmented into transient and non-transient portions using a transient detector. The transient portions are extracted and preserved separately, while the non-transient portions undergo frequency-dependent processing such as phase vocoding. This segmentation allows easy manipulation of the overall signal while protecting transient quality by excluding them from processing that would destroy vertical coherence.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Transient portions are extracted from the signal using a transient detector that identifies regions with rapidly changing energy distribution. These extracted transients are set aside and not subjected to frequency-dependent processing that would destroy their vertical coherence. After processing the remaining signal, the transients are reinserted to maintain their original high quality.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP2293294B1Device and method for manipulating an audio signal having a transient event
Publication Date: 2019.07.24 FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV
  • EP2293294B1 patent drawingFigure 1
  • EP2293294B1 patent drawingFigure 2
  • EP2293294B1 patent drawingFigure 3A~3B

AI summary

A signal manipulator for manipulating an audio signal having a transient event may comprise a transient remover (100), a signal processor (110) and a signal inserter (120) for inserting a time portion in a processed audio signal at a signal location where the transient event was removed before processing by said transient remover, so that a manipulated audio signal comprises a transient event not influenced by the processing, whereby the vertical coherence of the transient event is maintained instead of any processing performed in the signal processor (110), which would destroy the vertical coherence of a transient.