公开(公告)号：US20150149162A1

公开(公告)日：2015-05-28

申请号：US14087885

申请日：2013-11-22

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Ilya Dan MELAMED ,  Andrej LJOLJE

IPC: G10L15/00

CPC classification number: G10L15/07 ,  G10L15/20 ,  G10L15/28 ,  G10L2015/227

Abstract: Disclosed herein are systems, methods, and computer-readable storage devices for performing per-channel automatic speech recognition. An example system configured to practice the method combines a first audio signal of a first speaker in a communication session and a second audio signal from a second speaker in the communication session as a first audio channel and a second audio channel. The system can recognize speech in the first audio channel of the recording using a first model associated with the first speaker, and recognize speech in the second audio channel of the recording using a second model associated with the second speaker, wherein the first model is different from the second model. The system can generate recognized speech as an output from the communication session. The system can identify the models based on identifiers of the speakers, such as a telephone number, an IP address, a customer number, or account number.

Abstract translation: 这里公开了用于执行每通道自动语音识别的系统，方法和计算机可读存储装置。 配置为实施该方法的示例系统将通信会话中的第一扬声器的第一音频信号和来自通信会话中的第二扬声器的第二音频信号组合为第一音频通道和第二音频通道。 该系统可以使用与第一扬声器相关联的第一模型识别记录的第一音频通道中的语音，并且使用与第二扬声器相关联的第二模型识别记录的第二音频通道中的语音，其中第一模型不同 从第二个模型。 该系统可以将识别的语音产生为来自通信会话的输出。 该系统可以基于扬声器的标识符来识别模型，例如电话号码，IP地址，客户号码或帐号。