Method and system for sound source separation
AssigneeDublin Institute of Technology
InventorsDan Barry, Robert Lawlor, Eugene Coyle
Life of this patent
Estimating the term…
Abstract
Methods of sound source separation in which individual sources are extracted from a multiple source recording, include a method of analyzing stereo recordings to facilitate separation of individual musical sound sources from stereo music recordings. In the method sources predominant in the left are treated in a different manner to sources in the right.
Claims19
A method of modifying a stereo recording for subsequent analysis, the stereo recording comprising a first channel signal and a second channel signal, the method comprising the steps of:
- converting the first channel signal into the frequency domain,
- converting the second channel signal into the frequency domain,
- defining a set of scaling factors,
- producing a frequency azimuth plane by
- 1) gain scaling the frequency converted first channel signal by a first scaling factor selected from the set of defined scaling factors,
- 2) subtracting the gain scaled first channel signal from the frequency converted second channel signal,
- 3) repeating steps 1) and 2) individually for the remaining scaling factors in the defined set to produce the frequency azimuth plane, the frequency azimuth plane representing magnitudes of different frequencies for respective scaling factors and which can be used for subsequent analysis.
A sound analysis system comprising:
- an input module for accepting a first channel signal and a second channel signal,
- a first frequency conversion engine being adapted to convert the first channel signal into the frequency domain,
- a second frequency conversion engine being adapted to convert the second channel signal into the frequency domain, and
- a plane generator being adapted to gain scale the frequency converted first channel signal by a series of scaling factors from a previously defined set of scaling factors, subtract the gain scaled frequency converted first channel signals from the frequency converted second signal, and combining the resulting scale subtracted values to produce a frequency azimuth plane which represents magnitudes of different frequencies for each of the scaling.
A sound analysis system according to claim 14, wherein the plane generator is further adapted to gain scale the frequency converted second signal by a first scaling factor selected from the set of defined scaling factors and to subtract the gain scaled frequency converted second channel signal from the frequency converted first channel signal and to repeat this individually for the remaining scaling factors in the defined set and to combine the resulting gain scaled subtracted values with the previously determined gain scaled subtracted values in claim 18 to produce the frequency azimuth plane.
Dependencies are inferred from the claim text.