/Patent Tracker
AmazonGranted patentAudio & voiceInterest 6/10

Amazon patents speech system that works even when internet drops

Amazon describes a method where a local device, like a smart speaker, handles basic voice commands immediately to save time. It also sends the audio to a remote server for double-checking or to handle complex requests. If the connection to the server is lost, the device asks the user for more details to complete the task locally.

Why it mattersThis approach could make voice assistants feel faster and more reliable by reducing the need to wait for a cloud connection for simple commands.

In the filing’s own words

Speech processing tasks may be allocated at least partly to a local device (e.g., user computing device that receives spoken words) and at least partly to a remote device to determine one or more user commands or tasks to be performed by the local device. The remote device may be used to process speech that the local device could not process or understand, or for other reasons, such as for error checking. The local device may then execute or begin to execute locally determined tasks to reduce user-perceived latency. Meanwhile, the entire media input, or a portion thereof, may be sent to the remote device to process speech, verify the tasks and/or identify other user commands in the media input (or portion thereof).

What exactly they claim (claim 1)

1. A method comprising: receiving, at a first device, first audio data corresponding to a first user utterance of a user; generating, at the first device, first data by performing first speech recognition processing on the first audio data; sending, from the first device to a second device that is remote from the first device, the first data and results of the first speech recognition processing, the results being used by the second device to perform second speech recognition processing on at least one of the first data or the first audio data to determine a first task to be performed in response to the first user utterance; receiving, at the first device from the second device, second data corresponding to the first task; causing, at the first device, the first task to be performed; receiving, at the first device, second audio data corresponding to a second user utterance of a user; determining that the first device has lost connection with the second device; requesting, based on determining the first device has lost connection with the second device, additional information from the user; and performing a second task based on the additional information.

Facts

Company
Amazon (Amazon Technologies, Inc.)
Official title
Allocation of local and remote resources for speech processing
Patent number
US12731578B1
Application number
18217099
Filed
Jun 30, 2023
Granted
Sep 8, 2026
Inventors
2
Claims
18

Classified under

G10L 15/18 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Amazon in Audio & voice

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.