US 11,721,347 B1
Intermediate data for inter-device speech processing
Stanislaw Ignacy Pasko, Zawonia (PL); Pawel Zelazko, Gdansk (PL); Cagdas Bak, Gdansk (PL); Eli Joshua Fidler, Toronto (CA); Michal Kowalczuk, Gdansk (PL); Andrew Oberlin, Lynnwood, WA (US); and Ariya Rastrow, Seattle, WA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Jun. 29, 2021, as Appl. No. 17/362,301.
Int. Cl. G10L 17/26 (2013.01); G10L 15/183 (2013.01); G10L 15/34 (2013.01); G10L 15/22 (2006.01)
CPC G10L 17/26 (2013.01) [G10L 15/183 (2013.01); G10L 15/22 (2013.01); G10L 15/34 (2013.01)] 20 Claims
OG exemplary drawing
 
3. A computer-implemented method comprising:
receiving first audio data representing a first portion of an utterance;
performing first automatic speech recognition (ASR) processing on the first audio data using a first ASR component of a first device to generate first data representing a possible transcription of the first portion of the utterance;
sending the first data to a second device;
processing the first audio data to identify one or more characteristics of the first audio data;
sending second data representing the one or more characteristics to the second device;
performing second ASR processing on the first data using a second ASR component of the second device to determine a first ASR hypothesis corresponding to the first portion of the utterance;
performing first natural language understanding (NLU) processing on the first ASR hypothesis using an NLU component of the second device to generate first NLU results data; and
processing at least a portion of the first NLU results data using a skill component of the second device to perform a first action responsive to the first portion of the utterance.