公开(公告)号：WO2019124742A1

公开(公告)日：2019-06-27

申请号：PCT/KR2018/013821

申请日：2018-11-13

Applicant: 삼성전자 주식회사

Inventor： 한영호 ,  김남훈 ,  노재영 ,  박치연 ,  이경민 ,  조근석 ,  류종엽

IPC: G10L17/02 ,  G10L17/22 ,  G10L15/04 ,  G06F3/16 ,  G10L15/22

CPC classification number: G06F3/16 ,  G10L15/04 ,  G10L15/22 ,  G10L17/02 ,  G10L17/22

Abstract: 본 개시의 실시예에 따른 전자 장치는 메모리에 저장된 하나 이상의 인스트럭션을 실행함으로써, 수신부를 통해 음성 신호를 수신하도록 제어하고, 수신된 음성 신호에 서로 다른 복수 화자의 음성신호가 포함되어 있는지를 판단하고, 수신된 음성 신호에 서로 다른 복수 화자의 음성 신호가 포함되어 있으면, 각 화자의 음성 신호로부터 특징 정보를 검출하고, 검출된 특징 정보에 기초하여 서로 다른 화자의 발화 내용간의 관계를 판단하고, 판단된 발화 내용간의 관계에 기초하여 대응 방식을 결정하고, 결정된 대응 방식에 따라 전자 장치의 동작이 수행되도록 전자 장치를 제어하는 프로세서를 포함한다.

公开(公告)号：WO2019112332A1

公开(公告)日：2019-06-13

申请号：PCT/KR2018/015397

申请日：2018-12-06

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： JIN, Jangho

IPC: G10L15/30 ,  G10L25/51 ,  G10L15/22 ,  G06F3/16

CPC classification number: G10L15/22 ,  G10L15/30 ,  G10L17/22 ,  G10L2015/228

Abstract: An electronic apparatus is provided. The electronic apparatus includes a communicator comprising communication circuitry configured to communicate with a voice recognition server; and a processor configured to control the communicator to establish a session with the voice recognition server, based on a voice input start signal being received from a first external apparatus, to maintain the established session based on the voice input start signal being received from a second external apparatus in a state where the session is established, and to process voice recognition on audio data received from the second external apparatus using the maintained session.

公开(公告)号：WO2019027531A1

公开(公告)日：2019-02-07

申请号：PCT/US2018/032681

申请日：2018-05-15

Applicant: GOOGLE LLC

Inventor： SAK, Hasim ,  MORENO, Ignacio Lopez ,  PAPIR, Alan Sean ,  WAN, Li ,  WANG, Quan

IPC: G10L17/04 ,  G10L17/18 ,  G10L17/02 ,  G10L17/00

CPC classification number: G10L17/22 ,  G10L17/005 ,  G10L17/02 ,  G10L17/04 ,  G10L17/06 ,  G10L17/18

Abstract: Systems, methods, devices, and other techniques for training and using a speaker verification neural network. A computing device may receive data that characterizes a first utterance. The computing device provides the data that characterizes the utterance to a speaker verification neural network. Subsequently, the computing device obtains, from the speaker verification neural network, a speaker representation that indicates speaking characteristics of a speaker of the first utterance. The computing device determines whether the first utterance is classified as an utterance of a registered user of the computing device. In response to determining that the first utterance is classified as an utterance of the registered user of the computing device, the device may perform an action for the registered user of the computing device.

公开(公告)号：WO2019014425A1

公开(公告)日：2019-01-17

申请号：PCT/US2018/041779

申请日：2018-07-12

Applicant: PINDROP SECURITY, INC.

Inventor： PAYAS, Gupta ,  NELMS, Terry

IPC: H04L9/32 ,  H04L9/08 ,  H04L9/30

CPC classification number: H04L9/3231 ,  G10L17/06 ,  G10L17/22 ,  H04L9/0841 ,  H04L9/0869 ,  H04L9/3013 ,  H04L9/3066 ,  H04L9/3218

Abstract: Disclosed herein are embodiments of systems and methods for zero-knowledge multiparty secure sharing of voiceprints. In an embodiment, an illustrative computer may receive, through a remote server, a plurality of encrypted voiceprints. When the computer receives an incoming call, the computer may generate a plaintext i-vector of the incoming call. Using the plaintext i- vector and the encrypted voiceprints, the computer may generate one or more encrypted comparison models. The remote server may decrypt the encrypted comparison model to generate similarity scores between the plaintext i-vector and the plurality of encrypted voiceprints.

公开(公告)号：WO2017212235A1

公开(公告)日：2017-12-14

申请号：PCT/GB2017/051621

申请日：2017-06-06

Applicant: CIRRUS LOGIC INTERNATIONAL SEMICONDUCTOR LIMITED

Inventor： VAQUERO AVILÉS-CASCO, Carlos ,  MARTÍNEZ GONZÁLEZ, David ,  ROBERTS, Ryan

IPC: G10L17/22 ,  G06F21/32 ,  G10L15/08

CPC classification number: G10L17/22 ,  G06F21/32 ,  G10L17/005 ,  G10L17/02 ,  G10L17/04 ,  G10L2015/088

Abstract: A method of speaker authentication comprises: receiving a speech signal; dividing the speech signal into segments; and, following each segment, obtaining an authentication score based on said segment and previously received segments, wherein the authentication score represents a probability that the speech signal comes from a specific registered speaker. In response to an authentication request, an authentication result is output based on the authentication score.

Abstract translation: 说话人认证的方法包括：接收语音信号; 将语音信号分成段; 并且在每个片段之后，基于所述片段和先前接收的片段获得认证分数，其中认证分数表示语音信号来自特定注册讲话者的概率。 响应认证请求，根据认证分数输出认证结果。

公开(公告)号：WO2017131582A1

公开(公告)日：2017-08-03

申请号：PCT/SG2017/050008

申请日：2017-01-09

Applicant: MASTERCARD ASIA/PACIFIC PTE LTD

Inventor： PUEHSE, Tobias ,  ZHANG, Jie

IPC: G06Q30/06 ,  B25J5/00

CPC classification number: G06Q30/0613 ,  B25J11/0005 ,  B25J11/0015 ,  G06Q20/40145 ,  G06Q30/0635 ,  G10L13/00 ,  G10L15/08 ,  G10L15/22 ,  G10L17/06 ,  G10L17/08 ,  G10L17/22 ,  G10L17/26 ,  G10L2015/088 ,  G10L2015/223

Abstract: The present disclosure relates to methods for facilitating a transaction. The methods involve interaction with a humanoid robot and include using the humanoid robot to receive a session initiation instruction from a user. During the session one or more articles are identified for purchase, each article being a product or service. A checkout sequence is then initiated to purchase the one or more articles. At some stage during performance of the above steps, beforehand or afterwards, the user is authenticated via interaction with the humanoid robot. [FIG. 1]

Abstract translation: 本公开涉及用于促进交易的方法。 这些方法涉及与类人机器人的交互，并且包括使用人形机器人接收来自用户的会话发起指令。 在会议期间，确定购买一件或多件商品，每件商品为产品或服务。 结账顺序然后被启动以购买一个或多个物品。 在执行上述步骤的某个阶段，事先或之后，通过与仿人机器人的交互来验证用户。 [图。 1]

公开(公告)号：WO2017092322A1

公开(公告)日：2017-06-08

申请号：PCT/CN2016/089076

申请日：2016-07-07

Applicant: 乐视控股(北京)有限公司 ,  乐视致新电子科技(天津)有限公司

Inventor： 余绍鹏

IPC: H04N21/4782

CPC classification number: G10L15/1822 ,  G10L15/10 ,  G10L15/22 ,  G10L17/22 ,  G10L2015/223 ,  H04N21/42203 ,  H04N21/439 ,  H04N21/440236 ,  H04N21/4415 ,  H04N21/454 ,  H04N21/472 ,  H04N21/8173

Abstract: 本发明实施例提供一种智能电视的浏览器操作方法及智能电视，其中，所述方法包括：当所述智能电视的当前界面为所述浏览器的运行界面时，获取并解析来自所述智能电视的麦克风的声音数据；将解析后的声音数据与所述浏览器中预置的声音指令进行匹配；执行与所述解析后的声音数据匹配的声音指令相对应的浏览器操作。本发明实施例通过用户的声音来传达操作指令，避免了使用智能电视的遥控器对浏览器进行操作，提高了操作速度。

公开(公告)号：WO2017084197A1

公开(公告)日：2017-05-26

申请号：PCT/CN2016/070270

申请日：2016-01-06

Applicant: 深圳创维-RGB电子有限公司

Inventor： 付春元

IPC: G05B19/02 ,  G05B19/418 ,  G10L15/18

CPC classification number: G10L15/22 ,  G05B15/02 ,  G05B19/418 ,  G05B2219/2642 ,  G06F17/2735 ,  G06F17/2785 ,  G10L15/1815 ,  G10L15/26 ,  G10L17/06 ,  G10L17/22 ,  G10L25/63 ,  G10L2015/223

Abstract: 一种基于情感识别的智能家居控制方法及其系统，所述方法包括：获取用户的语音信息，对所述语音信息进行语音音调情感识别生成第一情感识别结果（S1）；将所述语音信息转换为文字信息后，对所述文字信息进行语义情感识别生成第二情感识别结果（S2）；基于所述第一情感识别结果和第二情感识别结果，根据预定的情感识别结果判断方法生成用户情感识别结果，并且根据所述用户情感识别结果，控制各智能家居设备执行相应的操作（S3）。通过分析用户当前的心情来自动控制家中的智能设备，并据此改变周围的环境条件，智能化程度较好。另外，采用语音语调识别和语义情感解析相结合的方法来进一步提升情感识别的准确性。

公开(公告)号：WO2017044370A1

公开(公告)日：2017-03-16

申请号：PCT/US2016/049856

申请日：2016-09-01

Applicant: VOICEBOX TECHNOLOGIES CORPORATION

Inventor： BRAGA, Daniela ,  ROMANI, Faraz ,  ELSHENAWY, Ahmad, Khamis ,  KENNEWICK, Michael

IPC: G10L15/26 ,  G10L15/06

CPC classification number: G10L15/30 ,  G06F3/0481 ,  G06F3/04842 ,  G06F3/0488 ,  G06F3/162 ,  G06Q10/10 ,  G10L15/00 ,  G10L15/01 ,  G10L15/063 ,  G10L17/22 ,  H04W4/21

Abstract: Systems and methods of providing text related to utterances, and gathering voice data in response to the text are provide herein. In various implementations, an identification token that identifies a first file for a voice data collection campaign, and a second file for a session script may be received from a natural language processing training device. The first file and the second file may be used to configure the mobile application to display a sequence of screens, each of the sequence of screens containing text of at least one utterance specified in the voice data collection campaign. Voice data may be received from the natural language processing training device in response to user interaction with the text of the at least one utterance. The voice data and the text may be stored in a transcription library.

Abstract translation: 本文提供了提供与话语相关的文本以及响应于文本收集语音数据的系统和方法。 在各种实施方式中，可以从自然语言处理训练装置接收识别用于语音数据收集活动的第一文件的识别令牌和用于会话脚本的第二文件。 可以使用第一文件和第二文件来配置移动应用程序来显示屏幕序列，每个屏幕序列包含语音数据收集活动中指定的至少一个话语的文本。 响应于用户与至少一个话语的文本交互而可以从自然语言处理训练装置接收语音数据。 语音数据和文本可以存储在转录库中。

公开(公告)号：WO2016164251A1

公开(公告)日：2016-10-13

申请号：PCT/US2016/025276

申请日：2016-03-31

Applicant: GOOGLE INC.

Inventor： CARBUNE, Victor ,  KEYSERS, Daniel M. ,  DESELAERS, Thomas

IPC: G10L15/26 ,  G06F17/24 ,  G10L15/193 ,  G10L25/48

CPC classification number: G10L15/26 ,  G06F17/243 ,  G10L15/193 ,  G10L17/22 ,  G10L25/48

Abstract: In some implementations, user input is received while a form that includes text entry fields is being accessed. In one aspect, a process may include mapping user input to fields of a form and populating the fields of the form with the appropriate information. This process may allow a user to fill out a form using speech input, by generating a transcription of input speech, determining a field that best corresponds to each portion of the speech, and populating each field with the appropriate information. In some examples, the processes described herein may reduce the load on user input components, may reduce overall power consumption and may reduce a cognitive burden on the user.

Abstract translation: 在一些实现中，接收用户输入，而正在访问包括文本输入字段的表单。 在一个方面，过程可以包括将用户输入映射到表单的字段并且用适当的信息填充表单的字段。 该过程可以允许用户使用语音输入来填写表单，通过生成输入语音的转录，确定与语音的每个部分最佳对应的字段，以及用适当的信息填充每个字段。 在一些示例中，本文描述的过程可以减少对用户输入组件的负担，可以降低总体功耗并且可以减少用户的认知负担。