US 12,444,407 B1
Privacy mode for multi-device processing
Fabian Andreas Bumberger, North York (CA); Eli Joshua Fidler, Toronto (CA); Maciej Grabon, Legnica (PL); Ashutosh Thakur, Toronto (CA); Ieliena Cannata, Wilmington, MA (US); and Muhammad Zain Mustafa, Kingston (CA)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Sep. 26, 2022, as Appl. No. 17/952,892.
Int. Cl. G10L 15/08 (2006.01); G10L 15/22 (2006.01); G10L 15/30 (2013.01)
CPC G10L 15/08 (2013.01) [G10L 15/22 (2013.01); G10L 15/30 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
capturing, by a first device, audio comprising an utterance;
determining, by the first device, audio data representing the utterance;
determining the first device is configured to operate in a first mode with respect to the utterance, the first mode corresponding to preventing audio recordings of a user's voice from leaving devices associated with a physical environment;
determining the first device is connected using a local area network protocol to a second device, the second device configured to perform computing functionality corresponding to the first mode, wherein the computing functionality is not configured for the first device;
sending, from the first device to the second device using the local area network protocol, the audio data;
performing automatic speech recognition (ASR) processing on the audio data by the second device to determine interim ASR results data;
sending, from the second device to the first device using the local area network protocol, the interim ASR results data; and
sending, from the first device to a third device over a wide area network, the interim ASR results data for further speech processing.