Detect triggering of digital assistants

Abstract

Systems and processes for operating an intelligent automated assistant are provided. According to one example, a method includes, at an electronic device having one or more processors, memory, and a plurality of microphones, sampling an audio signal at each of the plurality of microphones of the electronic device to Obtaining a plurality of audio signals; processing the plurality of audio signals to obtain a plurality of audio streams; and determining whether any of the plurality of audio signals corresponds to a voice trigger based on the plurality of audio streams. The method also includes initiating a session with the digital assistant based on determining that the plurality of audio signals correspond to the voice trigger; and aborting initiating the session based on determining that the plurality of audio signals do not correspond to the voice trigger. A conversation with a digital assistant.

Description

[0001] This application claims priority from the following patent applications: U.S. Provisional Patent Application No. 62 / 507,042, entitled "DETECTING A TRIGGER OF A DIGITAL ASSISTANT," filed May 16, 2017; Danish Patent Application No.PA201770420, titled "DETECTING A TRIGGER OF A DIGITAL ASSISTANT", filed May 31, 2017; and Danish Patent Application No. PA201770421, titled "DETECTING A TRIGGER OF A DIGITAL ASSISTANT" Triggering of Digital Assistants), submitted May 31, 2017. The content of the above patent application is hereby incorporated by reference in its entirety. technical field [0002] The present invention relates generally to intelligent automated assistants, and more particularly to detecting triggers of digital assistants. Background technique [0003] Intelligent automated assistants (or digital assistants) can provide a favorable interface between human users and electronic devices. Such assistants may allow users to interact with a device or system using na...

AI Technical Summary

Problems solved by technology

However, the sampled audio signal may include interfering audio signals, such as speech from competing speakers (e.g., people other than the user who is physically close to the electronic device), audio output from noise sources (e.g., a television in a room), mixed ring

As a result, the electronic device may fail to detect a voice trigger from the user due to interfering noise, or may falsely detect a voice trigger from an interfering audio signal (e.g., mistaking "History" from a competing speaker for "Hey Siri (Hi Siri)")

In addition, the electronic device may not be able to distinguish between portions of the sampled audio signal intended for the digital assistant and portions of the sampled audio signal not intended for the digital assistant (e.g., a conversation between two speakers)

This, in turn, leads to a confusing and inefficient user experience, as the electronic device may initiate a digital assistant session when the user does not issue a voice trigger, fail to initiate a digital assistant session when the user issues a voice trigger, and / or misinterpret user requests to the digital assistant

Images