US 11,676,598 B2
System and method for data augmentation for multi-microphone signal processing
Dushyant Sharma, Woburn, MA (US); Patrick A. Naylor, Reading (GB); Rong Gong, Vienna (AT); Stanislav Kruchinin, Vienna (AT); and Ljubomir Milanovic, Vienna (AT)
Assigned to Nuance Communications, Inc., Burlington, MA (US)
Filed by Nuance Communications, Inc., Burlington, MA (US)
Filed on May 7, 2021, as Appl. No. 17/314,660.
Claims priority of provisional application 63/022,269, filed on May 8, 2020.
Prior Publication US 2021/0352405 A1, Nov. 11, 2021
Int. Cl. H04R 3/00 (2006.01); H04R 3/04 (2006.01); G10L 15/22 (2006.01); H04R 1/40 (2006.01); G10L 25/84 (2013.01); G10L 15/32 (2013.01); G10L 15/20 (2006.01); G06F 16/65 (2019.01); G06F 16/68 (2019.01); G10L 17/06 (2013.01); G10L 25/78 (2013.01); H04R 5/04 (2006.01); H04S 7/00 (2006.01); H04R 29/00 (2006.01); G16H 15/00 (2018.01); G06N 20/00 (2019.01); G10L 21/028 (2013.01); G10L 15/26 (2006.01); G16H 10/60 (2018.01); G16H 40/20 (2018.01); G10L 21/0216 (2013.01); G10L 21/0272 (2013.01)
CPC G10L 15/22 (2013.01) [G06F 16/65 (2019.01); G06F 16/686 (2019.01); G06N 20/00 (2019.01); G10L 15/20 (2013.01); G10L 15/32 (2013.01); G10L 17/06 (2013.01); G10L 21/028 (2013.01); G10L 25/78 (2013.01); G10L 25/84 (2013.01); G16H 15/00 (2018.01); H04R 1/406 (2013.01); H04R 3/005 (2013.01); H04R 3/04 (2013.01); H04R 5/04 (2013.01); H04R 29/005 (2013.01); H04S 7/307 (2013.01); G10L 15/26 (2013.01); G10L 21/0216 (2013.01); G10L 21/0272 (2013.01); G10L 2021/02166 (2013.01); G16H 10/60 (2018.01); G16H 40/20 (2018.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, executed on a computing device, comprising:
receiving a signal from each microphone of a plurality of microphones, thus defining a plurality of signals;
receiving one or more microphone frequency responses associated with at least one microphone;
performing one or more microphone frequency response-based augmentations on the plurality of signals based upon, at least in part, the one or more microphone frequency responses, thus defining one or more microphone frequency response-based augmented signals; and
training one or more models representative of a microphone frequency response based upon, at least in part, the one or more microphone frequency response-based augmentations on the plurality of signals.