公开(公告)号：US20150310853A1

公开(公告)日：2015-10-29

申请号：US14261650

申请日：2014-04-25

Applicant: GM GLOBAL TECHNOLOGY OPERATIONS LLC

Inventor： CODY R. HANSEN ,  TIMOTHY J. GROST ,  UTE WINTER

IPC: G10L15/08

CPC classification number: G10L15/08 ,  G10L15/20 ,  G10L15/22 ,  G10L21/0364

Abstract: A method for speech recognition includes generating a speech prompt, receiving a spoken utterance from a user in response to the speech prompt, wherein the spoken utterance includes a speech artifact, and compensating for the speech artifact. Compensating for the speech artifact may include, for example, utilizing a recognition grammar that includes the speech artifact as a speech component, or modifying the spoken utterance to eliminate the speech artifact.

Abstract translation: 用于语音识别的方法包括产生语音提示，响应于语音提示接收来自用户的口语话语，其中所述语音话语包括语音伪像，并补偿语音伪像。 对于语音伪像的补偿可以包括例如利用包括语音伪像作为语音分量的识别语法，或修改语音话语以消除语音伪影。

公开(公告)号：US20150317973A1

公开(公告)日：2015-11-05

申请号：US14266593

申请日：2014-04-30

Applicant: GM Global Technology Operations LLC

Inventor： CODY R. HANSEN ,  ROBERT A. HRABAK ,  TIMOTHY J. GROST

IPC: G10L15/08 ,  G10L17/22 ,  G10L15/22

CPC classification number: G10L15/08 ,  G10L15/1822 ,  G10L15/22 ,  G10L15/32 ,  G10L17/22

Abstract: Methods and systems are provided for coordinating recognition of a speech utterance between a speech system of a vehicle and a speech system of a user device. In one embodiment, a method includes: receiving the speech utterance from a user; performing speech recognition on the speech utterance to determine a topic of the speech utterance; determining whether the speech utterance was meant for the speech system of the vehicle or the speech system of the user device based on the topic of the speech utterance; and selectively providing the speech utterance to the speech system of the vehicle or the speech system of the user device based on the determination of whether the speech utterance was meant for the speech system of the vehicle or the speech system of the user device.

Abstract translation: 提供了用于协调对车辆的语音系统和用户设备的语音系统之间的语音发声的识别的方法和系统。 在一个实施例中，一种方法包括：从用户接收语音话语; 对演讲话语进行语音识别，以确定讲话话语的主题; 基于语音话语的话题来确定语音话语是否用于车辆的语音系统或用户设备的语音系统; 并且基于确定所述语音语音是否用于所述车辆的语音系统或所述用户设备的语音系统的所述确定来选择性地向所述车辆的语音系统或所述用户设备的所述语音系统提供语音发声。