US 12,073,837 B2
Speech detection using image classification
Stephen Gregory Dame, Everett, WA (US); and Les Eugene Atlas, Seattle, WA (US)
Assigned to The Boeing Company, Arlington, VA (US); and University of Washington, Seattle, WA (US)
Filed by The Boeing Company, Chicago, IL (US); and University of Washington, Seattle, WA (US)
Filed on Jun. 7, 2022, as Appl. No. 17/805,822.
Claims priority of provisional application 63/202,659, filed on Jun. 18, 2021.
Prior Publication US 2022/0406310 A1, Dec. 22, 2022
Int. Cl. G10L 15/24 (2013.01); G06V 10/82 (2022.01); G10L 15/05 (2013.01)
CPC G10L 15/24 (2013.01) [G06V 10/82 (2022.01); G10L 15/05 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method performed by a computing system, the method comprising:
obtaining an audio segment of radio communications;
extracting an audio sub-segment within the audio segment;
generating a sampled histogram of a plurality of sampled values across a sampled time window of the audio sub-segment;
generating a two-dimensional image that represents a two-dimensional mapping of the sampled histogram along a first dimension and a predefined histogram along a second dimension that is orthogonal to the first dimension;
providing the two-dimensional image to an image classifier previously trained using the predefined histogram; and
receiving an output from the image classifier based on the two-dimensional image, the output indicating whether the audio sub-segment contains speech.