Abstract

Devices that provide voice assistant features can handle voice commands or spoken queries locally on device or by offloading the processing to an external server. Local query processing can eliminate the latency and potential unreliability of server-side processing. When local processing and fulfillment is infeasible or unlikely, query fulfillment is performed on a server, if permitted by the user. In such a case, the computation involved in the unsuccessful attempt at local processing consumes local resources without a usable result. This disclosure describes techniques to determine locally if a spoken query cannot be served by the device and abandoning local processing upon such a determination. The user is informed of the query abandonment, and if the user permits, server-side query processing is carried out to provide a response to the query.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.

Share

COinS