Active listening for assistant system

In one embodiment, a method includes receiving a first user input from a first client system, the first user input including a wake-up word associated with an assistant xbottom; setting the assistant xpot to a listening mode, wherein continuous non-visual feedback is provided via the first client sy...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: CHEN YUANHUI, WIGDOR DANIEL JOHN, JUNJUNWALA, MIKA, BELLMAN AMY LAWSON, FLORES NICHOLAS JORGE, STEWART RYAN FREDERICK, GAN XIN, LIU BING, HILAIRE LLOYD, VENKATESH, GANESH, DABAS, LUOXIN, LIU BAIYANG, HANSEN MICHAEL ROBERT, CUI ZHENHUA, MOON, SEUNG HWAN, ZHOU HAO, KITCHENS JESSICA, LEVISON, DAVID, CHALLAND CHRISTOPHE, BALMES CHRISTOPHER E, GURUNATH PRAMOD, GRAVES IAN, BLACK HEATH WILLIAM, ZUO ZHENGPING, YU JINSONG, NOTKE SAMUEL STEELE, SAVENKOV DENIS, DANE, JUSTIN, DE PAOLI, CHRISTOPHE, DILLAFSON, ALIREZA, SURKOV ALEXEY GENNADIEVICH, ARCH, KYLE, XU HU, GLUCK MICHAEL, NORTHUP ERIC ROBERT, BLAKELY JOHN JACOB, MARTINSEN LEIF HAVEN, SHALOWITZ, ILANA, ORLY, SRINIVAS, KRISHNA, CHAITANYA, GOPISETTY, MOSKEY GABRIEL CATHERINE, SETHI, PUJA, PARENT MARK, GOEL, SWATI, KAHN JEREMY GILMORE, PU YIMING, LIU HONGLEI, MATTHEW DAN FAZLEY, TIWARI MEGA, SANTOSA, STEPHANIE, RAMANAN, TARA, KAMKAR, PIYUSH, MOHAMED AHMED MAGADI HAMID, LI JIHANG, VINCENT JOSHUA, RACINE JACKSON
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In one embodiment, a method includes receiving a first user input from a first client system, the first user input including a wake-up word associated with an assistant xbottom; setting the assistant xpot to a listening mode, wherein continuous non-visual feedback is provided via the first client system while the assistant xpot is in the listening mode; receiving a second user input from the first client system while the assistant xboot is in a listening mode, the second user input including a user utterance; determining that the second user input has ended based on completion of the user utterance; and setting the assistant xpot to an inactive mode, where the non-visual feedback is terminated via the first client system while the assistant xpot is in the inactive mode. 在一个实施例中,一种方法包括:从第一客户端系统接收第一用户输入,该第一用户输入包括与助理xbot相关联的唤醒词;将助理xbot设置为收听模式,其中在该助理xbot处于收听模式时经由第一客户端系统提供连续非视觉反馈;在助理xbot处于收听模式时从第一客户端系统接收第二用户输入,该第二用户输入包括用户话语;基于用户话语的完成来确定第二用户输入已结束;以及将助理xbot设置为非活动模式,其中在助理xbot处于非活动模式时,经由第一客户端系统终止非视觉反馈。