US 12,334,104 B2
Method and system for time and feature modification of signals
Jonathan Newland, Epsom (GB); Phillip Jeffrey Bloom, Epsom (GB); and William John Ellwood, Epsom (GB)
Assigned to SYNCHRO ARTS LIMITED, Epsom (GB)
Filed by SYNCHRO ARTS LIMITED, London (GB)
Filed on Jul. 28, 2022, as Appl. No. 17/875,816.
Claims priority of application No. 2110883 (GB), filed on Jul. 28, 2021.
Prior Publication US 2023/0032838 A1, Feb. 2, 2023
Int. Cl. G10L 25/51 (2013.01); G10L 25/78 (2013.01); G11B 27/10 (2006.01)
CPC G10L 25/51 (2013.01) [G10L 25/78 (2013.01); G11B 27/10 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A computer implemented method for modifying one or more features of an input audio signal based on one or more features in a guide audio signal, the method comprising:
dividing the guide audio signal and the input audio signal into a plurality of time frames;
analysing the plurality of time frames in the guide audio signal and the plurality of time frames in the input audio signal;
classifying the plurality of time frames into time frames with a sufficient signal level and time frames with an insufficient signal level;
determining a matchability signal by comparing one or more audio features of the plurality of time frames with a sufficient signal level in the guide audio signal with one or more corresponding features of the plurality of corresponding time frames with a sufficient signal level in the input audio signal, the matchability signal indicating matchable sections of the input audio signal having audio features that can be modified to match the guide audio signal, and unmatchable sections of the input audio signal having audio features that cannot be modified to match the guide audio signal;
labelling each of the plurality of time frames as a matchable or unmatchable section of the input audio signal based at least in part on the matchability signal for each of the plurality of time frames;
relabelling any unmatchable section of the input audio signal as a matchable section of the input audio signal where the unmatchable section of the input audio signal has a length of time that is less than a pre-set minimum length;
filtering any unmatchable section of the input audio signal where the unmatchable section of the input audio signal has a length of time that is greater than the pre-set minimum length; and
modifying one or more audio features in the matchable sections of the input audio signal.