WO2017059815A1 - 一种快速识别方法及家庭智能机器人 - Google Patents

一种快速识别方法及家庭智能机器人 Download PDF

Info

Publication number
WO2017059815A1
WO2017059815A1 PCT/CN2016/101567 CN2016101567W WO2017059815A1 WO 2017059815 A1 WO2017059815 A1 WO 2017059815A1 CN 2016101567 W CN2016101567 W CN 2016101567W WO 2017059815 A1 WO2017059815 A1 WO 2017059815A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
robot
home
identification method
facial image
Prior art date
Application number
PCT/CN2016/101567
Other languages
English (en)
French (fr)
Inventor
向文杰
朱磊
Original Assignee
芋头科技(杭州)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 芋头科技(杭州)有限公司 filed Critical 芋头科技(杭州)有限公司
Priority to US15/766,890 priority Critical patent/US10664511B2/en
Priority to JP2018517885A priority patent/JP6620230B2/ja
Priority to EP16853119.2A priority patent/EP3361410A4/en
Publication of WO2017059815A1 publication Critical patent/WO2017059815A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/70Multimodal biometrics, e.g. combining information from different biometric modalities
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
    • G06F16/43Querying
    • G06F16/432Query formulation
    • G06F16/433Query formulation using audio data
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/907Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/10Terrestrial scenes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/161Detection; Localisation; Normalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/161Detection; Localisation; Normalisation
    • G06V40/166Detection; Localisation; Normalisation using acquisition arrangements
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172Classification, e.g. identification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/22Indexing; Data structures therefor; Storage structures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L17/00Speaker identification or verification

Definitions

  • the invention relates to the field of robots, in particular to a rapid identification method applied to a home intelligent robot and a home intelligent robot.
  • smart electronic products also known as home smart robots.
  • a rapid identification method for home smart robots including:
  • Step S100 preset a plurality of personal files corresponding to different users
  • Step S200 Collect identification information associated with the feature of the user, where the identification information is Establishing an association with the corresponding personal file of the user;
  • Step S300 the home intelligent robot collects the feature of the user and matches the stored identification information to identify the user;
  • step S400 is performed, otherwise, exiting;
  • Step S400 Retrieving the corresponding personal profile according to the identified user, and working according to the personal profile.
  • the fast identification method described above wherein the user activates the home smart robot by activating a voice and sends an instruction to the home smart robot.
  • the fast identification method described above wherein the identification information comprises a voiceprint model.
  • the fast identification method described above wherein the identification information comprises a facial image model.
  • the method for quickly identifying the voiceprint model includes a first active acquisition and a first automatic acquisition;
  • the first active acquisition pre-collects the activation voice of the user according to the home intelligence robot, and obtains the voiceprint model of the user;
  • the first automatic acquisition automatically obtains the voiceprint model of the user according to the activation voice used by the user for the first time collected by the home intelligence robot.
  • the method for quickly identifying the facial image model includes the second active acquisition and the second automatic acquisition;
  • the second active acquisition pre-collects a facial image image of the user according to the home smart robot to obtain the facial image model of the user;
  • the second automatic acquisition automatically reads the facial image image of the user after acquiring the voiceprint model of the user according to the home intelligent robot, and obtains the facial image model.
  • the quick identification method described above wherein the personal profile includes a history record and a favorite list
  • the home smart robot receiving an instruction of the identified user, according to the history in the personal profile of the identified user
  • the record or the list of favorites executes the instructions.
  • the fast identification method described above wherein a storage unit is provided for storing a plurality of time-related pre-recorded voices, the personal profile includes a name of the user, and the home intelligent robot automatically performs the user Facial image recognition, the name in the profile of the user is retrieved according to the recognition result, and the corresponding pre-recorded voice is selected in the storage unit according to the current time, and the name is sounded through the machine
  • the pre-recorded voice is spliced and played.
  • a camera is provided to read a facial image image of the user.
  • a home smart robot is further included, wherein the fast identification method described above is employed.
  • the invention has the beneficial effects that the user can be quickly identified, the recognition degree and the recognition rate are improved, the home robot is more intelligent, and the personalized service can be provided according to different users, which has broad application prospects.
  • FIG. 1 is a flow chart of a method for quickly identifying a method and a method for a home intelligent robot according to the present invention.
  • a rapid identification method suitable for a home intelligent robot wherein, as shown in FIG. 1, comprising:
  • Step S100 preset a plurality of personal files corresponding to different users
  • each user's personal profile is created to store relevant information of different users, such as according to favorite music or personal preferences, so that the home smart robot can provide different personalized services for each family member. .
  • Step S200 Collect identification information associated with a feature of the user, and establish an association between the identification information and a personal file of the corresponding user;
  • the user profile and the identification information are related to each other, and the corresponding user profile is obtained by the identification information, so that the home intelligence robot can work according to different information recorded in each personal file, thereby providing individuality for each family member. Service.
  • the identification information comprises a voiceprint model.
  • the acquisition mode of the voiceprint model includes a first active acquisition and a first automatic acquisition
  • the first active acquisition pre-acquires the user's activation voice according to the home intelligence robot to obtain the user's voiceprint model
  • Active acquisition is mainly for the first time when using the home intelligent robot, it needs to input some information to the robot and make settings for future use, such as setting the content of the activated voice, and collecting the identification information of each member of the family.
  • the first automatic acquisition automatically obtains the voiceprint model of the user according to the activation voice used by the user collected by the home intelligent robot for the first time.
  • the collection of identification information for new users can be automatically collected by the home intelligent robot. For example, when a new user issues an instruction to the home intelligent robot for the first time (such as calling a name set for the home intelligent robot), the home intelligent robot is activated according to the activation. The voice is activated, and the user's voice is collected, and the voiceprint model is produced, and the identification information is collected while responding to the user's instruction. And establish a personal file of the new user, and save the collected voiceprint model as identification information.
  • the identification information includes a facial image model.
  • the method for collecting the facial image model includes a second active acquisition and a second automatic acquisition;
  • the second active acquisition pre-collects the facial image of the user according to the home intelligence robot to obtain a facial image model of the user;
  • Active acquisition is mainly for the first time when using the home intelligent robot, it needs to input some information to the robot and make settings for future use, such as setting the content of the activated voice, and collecting the identification information of each member of the family.
  • the second automatic acquisition automatically reads the facial image of the user after the user's voiceprint model is acquired, and obtains a facial image model.
  • the facial image model is also collected for the new user, which is convenient for the user to identify the next time when using the home intelligent robot.
  • Step S300 The home intelligence robot collects the characteristics of the user and matches the stored identification information to identify the user;
  • step S400 is performed, otherwise, exiting;
  • the home intelligent robot recognizes the user, if the collected facial image blur cannot be used for facial recognition, the voiceprint of the user is automatically recognized, and if the user's identity is recognized by the voiceprint, Even if facial image recognition is not successful, the user can be recognized by the home smart robot by voice.
  • the smart home robot successfully identifies the user, and only when the face recognition and the voiceprint recognition are not successful, the smart home robot pairs The user's identification fails, and the user can identify the identity again by voice or facial image.
  • Step S400 according to the identified user, the corresponding personal file is retrieved, and the work is performed according to the personal file. Work.
  • the user activates the home smart robot by activating the voice and sends an instruction to the home smart robot.
  • the activated home smart robot is activated by using fixed voice, such as giving the home smart robot a good name, calling to the family, calling out
  • the name of the family intelligent robot is activated by the previous setting when the home intelligent robot hears its own name.
  • voiceprint recognition based on the activated voice can be made when the user uses the robot.
  • the robot is activated by issuing an activation voice, and when the robot detects a sound containing its own name, the voiceprint detection is performed, and therefore, the voiceprint detection based on the fixed voice has a high accuracy.
  • the personal profile includes a history record and a list of favorites
  • the home intelligence robot receives the command of the identified user and executes the command based on the history or favorite list in the profile of the identified user.
  • the robot can recognize the user by activating the voice, record the user's playlist, and analyze it. After the user uses the robot for a period of time, the robot can pass the The user's history and favorite list are accurately recommended.
  • the robot can distinguish family members through voiceprints and recommend different music for different family members.
  • a storage unit is further configured to store a plurality of pre-recorded voices associated with time.
  • the personal profile may further include a name of the user, and the home intelligent robot automatically performs facial image recognition on the user, and adjusts according to the recognition result.
  • the name in the user's profile is taken, and the corresponding pre-recorded voice is selected in the storage unit according to the current time, and the name is spliced through the machine sounding and pre-recorded voice and played.
  • the pre-recorded voice associated with the time is saved in the storage unit for voice broadcast when needed, for example, when the user returns home at night, the robot detects the person through the infrared camera device. Active self-activation, and identify the current user's identity through the facial image to obtain the user's personal profile, and obtain the corresponding pre-recorded voice in the storage unit according to the current time.
  • the home intelligent robot can pass the built-in TTS (Text To The Speech engine plays the name of the machine-sounding profile and stitches the acquired pre-recorded voice to form a greeting such as "Good evening, xxx" or plays the music that the user likes based on the history in the profile.
  • the greeting content may be stored in a storage unit as a character string, and the machine sounds directly through the TTS engine, thereby reducing the storage space required by the storage unit.
  • a camera is provided to read a facial image of a user.
  • the camera While detecting the voiceprint, the camera synchronously detects the user's face. If the user's facial image is not detected, the voiceprint data is separately saved; if the user's facial image is detected, the user's face and voiceprint data are simultaneously saved and associated with the personal file. And through interaction and user confirmation, you can establish the relationship between voiceprints, facial images, and personal files.
  • a home intelligent robot is further included, which adopts the above-described fast identification method.
  • the voiceprint model can be recognized at the same time or recognized by the face model, and the identification of multiple modes is beneficial to improve the accuracy and efficiency of the recognition. If the user activates the robot to activate the interaction, the voiceprint recognition can accurately identify the user; if the user does not use the active voice, the user can also be identified through the face.

Abstract

一种快速识别方法及家庭智能机器人,涉及智能电子领域。适用于家庭智能机器人,步骤S100、预先设置多个对应不同的用户的个人档案(S100);步骤S200、采集与用户的特征相关联的识别信息,于识别信息与对应的用户的个人档案之间建立关联(S200);步骤S300、家庭智能机器人采集用户的特征与存储的识别信息进行匹配以识别用户;如识别成功,执行步骤S400,否则,退出(S300);步骤S400、根据识别的用户调取对应的个人档案,根据个人档案进行工作(S400)。本方法的有益效果是:能够对不同的用户进行快速识别,提高识别度和识别率,使家庭机器人变得更加智能,并能根据不同的用户提供个性化服务,具有广阔的适用前景。

Description

一种快速识别方法及家庭智能机器人 技术领域
本发明涉及机器人领域,尤其涉及一种应用于家庭智能机器人的快速识别方法及家庭智能机器人。
背景技术
随着智能电子的普及,越来越多的家庭开始使用智能电子产品也称为家庭智能机器人。
目前智能化电子设备得到越来越多的使用,其中家庭智能机器人也逐步进入普通百姓的家中。但是现有的家庭智能机器人通常延续了传统移动终端的一对一操作模式,将家庭中所有成员作为一个用户进行对待,并不加以区分,从而令家庭中的每个成员要操作家庭智能机器人时无法获得专属的个性化服务,使家庭智能机器人的地位被弱化,从而影响了家庭智能机器人的推广。
发明内容
针对上述问题现提供能够对不同用户进行快速识别家庭成员并提供个性化服务的一种快速识别方法及家庭智能机器人。
具体的技术方案是:
一种快速识别方法,适用于家庭智能机器人,其中,包括:
步骤S100、预先设置多个对应不同的用户的个人档案;
步骤S200、采集与所述用户的特征相关联的识别信息,于所述识别信息 与对应的所述用户的所述个人档案之间建立关联;
步骤S300、所述家庭智能机器人采集所述用户的特征与存储的所述识别信息进行匹配以识别所述用户;
如识别成功,执行步骤S400,否则,退出;
步骤S400、根据识别的所述用户调取对应的所述个人档案,根据所述个人档案进行工作。
优选的,上述的快速识别方法,其中,所述用户通过激活语音启动所述家庭智能机器人,并向所述家庭智能机器人发送指令。
优选的,上述的快速识别方法,其中,所述识别信息包括声纹模型。
优选的,上述的快速识别方法,其中,所述识别信息包括面部影像模型。
优选的,上述的快速识别方法,其中,所述声纹模型的采集方式包括第一主动采集和第一自动采集;
所述第一主动采集根据所述家庭智能机器人预先采集所述用户的所述激活语音,获得所述用户的所述声纹模型;
所述第一自动采集根据所述家庭智能机器人采集的所述用户第一次使用的所述激活语音,自动获得所述用户的所述声纹模型。
优选的,上述的快速识别方法,其中,所述面部影像模型的采集方式包括第二主动采集和第二自动采集;
所述第二主动采集根据所述家庭智能机器人预先采集所述用户的面部影像图形,获得所述用户的所述面部影像模型;
所述第二自动采集根据所述家庭智能机器人在获取了所述用户的所述声纹模型后,自动读取所述用户的面部影像图形,获得所述面部影像模型。
优选的,上述的快速识别方法,其中,所述个人档案包括历史记录和收藏列表,所述家庭智能机器人接收已识别用户的指令,根据所述已识别用户的所述个人档案中的所述历史记录或所述收藏列表执行所述指令。
优选的,上述的快速识别方法,其中,提供一存储单元用以存储复数个与时间关联的预录制语音,所述个人档案包括所述用户的姓名,所述家庭智能机器人对所述用户自动进行面部影像识别,根据识别结果调取所述用户的所述个人档案中的所述姓名,并根据当前时间于所述存储单元中选择对应的所述预录制语音,将所述姓名通过机器发声与所述预录制语音拼接后播放。
优选的,上述的快速识别方法,其中,提供一摄像头对所述用户的面部影像图像进行读取。
优选的,还包括一种家庭智能机器人,其中,采用上述所述的快速识别方法。
本发明的有益效果是:能够对不同的用户进行快速识别,提高识别度和识别率,使家庭机器人变得更加智能,并能根据不同的用户提供个性化服务,具有广阔的适用前景。
附图说明
图1为本发明一种快速识别方法及家庭智能机器人的方法实施例流程图。
具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动的前提下所获得的所有其他实施例,都属于本发明保护的范围。
需要说明的是,在不冲突的情况下,本发明中的实施例及实施例中的特征可以相互组合。
下面结合附图和具体实施例对本发明作进一步说明,但不作为本发明的限定。
一种快速识别方法,适用于家庭智能机器人,其中,如图1所示,包括:
步骤S100、预先设置多个对应不同用户的个人档案;
本发明较佳实施例中,建立每个用户的个人档案,用以存储不同用户的相关信息如根据喜欢的音乐或个人爱好等,使家庭智能机器人可以为每个家庭成员提供不同的个性化服务。
步骤S200、采集与用户的特征相关联的识别信息,于识别信息与对应的用户的个人档案之间建立关联;
对用户个人档案与识别信息之间进行相互的关联,通过识别信息获得对应用户个人档案,使得家庭智能机器人能根据每个个人档案中记载的不同信息进行工作,从而实现为每个家庭成员提供个性化服务。
本发明较佳实施例中,识别信息包括声纹模型。
在本技术发明的基础上,进一步的,声纹模型的采集方式包括第一主动采集和第一自动采集;
第一主动采集根据家庭智能机器人预先采集用户的激活语音,获得用户的声纹模型;
主动采集主要针对第一次使用家庭智能机器人时,需要对机器人进行一些信息的录入以及对日后使用所做得设定,如设定激活语音的内容,以及家庭各个成员的识别信息采集。第一自动采集根据家庭智能机器人采集的用户第一次使用的激活语音,自动获得用户的声纹模型。
对于新用户的识别信息的采集可以通过家庭智能机器人的自动采集方式,比如新用户在第一次向家庭智能机器人发出指令时(如呼唤对家庭智能机器人设定的名字),家庭智能机器人根据激活语音被激活,并采集该用户的声音,生产声纹模型,在响应该用户的指令的同时,完成识别信息的采集 并建立新用户的个人档案,将采集的声纹模型作为识别信息进行保存。
本发明较佳实施例中,识别信息包括面部影像模型。
在本技术发明的基础上,进一步的,面部影像模型的采集方式包括第二主动采集和第二自动采集;
第二主动采集根据家庭智能机器人预先采集用户的面部影像图形,获得用户的面部影像模型;
主动采集主要针对第一次使用家庭智能机器人时,需要对机器人进行一些信息的录入以及对日后使用所做得设定,如设定激活语音的内容,以及家庭各个成员的识别信息采集。
第二自动采集根据家庭智能机器人在获取了用户的声纹模型后,自动读取用户的面部影像图形,获得面部影像模型。
对于新用户的识别信息进行自动采集时,也包括对新用户采集面部影像模型,方便用户下次使用家庭智能机器人时的身份识别。
步骤S300、家庭智能机器人采集用户的特征与存储的识别信息进行匹配以识别用户;
如识别成功,执行步骤S400,否则,退出;
本发明较佳实施例中,如家庭智能机器人对用户进行识别时,如采集的面部影像模糊无法进行面部识别,则自动对用户的声纹进行识别,如果通过声纹识别出用户的身份,则即使面部影像识别没有成功,通过语音,用户一样可被家庭智能机器人所识别。
作为进一步的优选实施方式,只要影像识别,或者声纹识别两者任一识别成功,则智能家庭机器人对用户的身份识别成功,只有在面部识别和声纹识别都没有成功时,智能家庭机器人对用户的识别失败,用户可以再次通过语音或面部影像进行身份的识别。
步骤S400、根据识别的用户调取对应的个人档案,根据个人档案进行工 作。
本发明较佳实施例中,用户通过激活语音启动家庭智能机器人,并向家庭智能机器人发送指令。
当用户对家庭智能机器人下指令时,为和用户其他的语言进行区分,一般对激活家庭智能机器人采用固定语音进行激活,如给家庭智能机器人起一个好听的名字,向对家人呼唤一样,叫出家庭智能机器人的名字,则通过前期的设置,当家庭智能机器人听到自己的名字时被激活,因为机器人的激活语音是固定的,可以做基于激活语音的声纹识别,当用户使用机器人的时候,通过发出激活语音来激活机器人,当机器人检测到包含自己名字的声音时,做声纹检测,因此,基于固定语音的声纹检测有较高的准确率。
本发明较佳实施例中,个人档案包括历史记录和收藏列表,家庭智能机器人接收已识别用户的指令,根据已识别用户的个人档案中的历史记录或收藏列表执行指令。
如,当某个用户激活机器人下指令说“帮我放音乐”,机器人可以通过激活语音来识别用户,记录该用户的播放列表,并进行分析,当用户使用一段时间以后,机器人就可以通过该用户的历史记录和收藏列表进行精准推荐,当另外一个家庭成员也下达了同样指令“帮我放音乐”,机器人就可以通过声纹来区分家庭成员,为不同的家庭成员推荐不同的音乐。
本发明较佳实施例中,还提供一存储单元用以存储复数个与时间关联的预录制语音,个人档案还可包括用户的姓名,家庭智能机器人对用户自动进行面部影像识别,根据识别结果调取用户的个人档案中的姓名,并根据当前时间于存储单元中选择对应的预录制语音,将姓名通过机器发声与预录制语音拼接后播放。
将与时间关联的预录制语音保存在存储单元中,用以在需要时进行语音播报,如,当用户晚上回家,机器人通过红外线的摄像设备检测到有人,可 以主动自我激活,并且通过面部影像识别当前用户的身份以获取用户的个人档案,并根据当前的时间于存储单元中获取对应的预录制语音,此时家庭智能机器人可通过内置的TTS(Text To Speech)引擎播放机器发声的个人档案中的姓名,并拼接获取的预录制语音,以形成如“晚上好,xxx”的问候语,或者根据在个人档案中的历史记录播放该用户喜欢的音乐。作为优选的实施方式,也可将问候语内容以字符串形式保存于存储单元中,直接通过TTS引擎进行机器发声,从而减小存储单元所需的存储空间。
本发明较佳实施例中,提供一摄像头对用户的面部影像图像进行读取。
在检测声纹的同时,摄像头同步检测用户面部,如果没有检测到用户面部影像,则单独保存声纹数据;如果检测到用户面部影像,将用户面部和声纹数据同时保存并与个人档案建立关联,并通过互动经过用户确认,就可以把声纹,面部影像,个人档案之间的关系建立起来。
本发明较佳实施例中,还包括一种家庭智能机器人,采用上述的快速识别方法。
当需要对用户进行身份识别时,可同时进行声纹模型进行识别或通过面部模型进行识别,多种方式的识别有利于提高识别的准确率和效率。如果用户通过激活语音激活机器人进行互动的时候,通过声纹识别可以准确的识别用户;如果用户没有使用激活语音,同样可以通过面部来识别用户。
以上所述仅为本发明较佳的实施例,并非因此限制本发明的实施方式及保护范围,对于本领域技术人员而言,应当能够意识到凡运用本发明说明书及图示内容所作出的等同替换和显而易见的变化所得到的方案,均应当包含在本发明的保护范围内。

Claims (10)

  1. 一种快速识别方法,适用于家庭智能机器人,其特征在于,包括:
    步骤S100、预先设置多个对应不同的用户的个人档案;
    步骤S200、采集与所述用户的特征相关联的识别信息,于所述识别信息与对应的所述用户的所述个人档案之间建立关联;
    步骤S300、所述家庭智能机器人采集所述用户的特征与存储的所述识别信息进行匹配以识别所述用户;
    如识别成功,执行步骤S400,否则,退出;
    步骤S400、根据识别的所述用户调取对应的所述个人档案,根据所述个人档案进行工作。
  2. 如权利要求1所述的快速识别方法,其特征在于,所述用户通过激活语音启动所述家庭智能机器人,并向所述家庭智能机器人发送指令。
  3. 如权利要求1所述的快速识别方法,其特征在于,所述识别信息包括声纹模型。
  4. 如权利要求1所述的快速识别方法,其特征在于,所述识别信息包括面部影像模型。
  5. 如权利要求3所述的快速识别方法,其特征在于,所述声纹模型的采集方式包括第一主动采集和第一自动采集;
    所述第一主动采集根据所述家庭智能机器人预先采集所述用户的所述激活语音,获得所述用户的所述声纹模型;
    所述第一自动采集根据所述家庭智能机器人采集的所述用户第一次使用的所述激活语音,自动获得所述用户的所述声纹模型。
  6. 如权利要求4所述的快速识别方法,其特征在于,所述面部影像模型的采集方式包括第二主动采集和第二自动采集;
    所述第二主动采集根据所述家庭智能机器人预先采集所述用户的面部影像图形,获得所述用户的所述面部影像模型;
    所述第二自动采集根据所述家庭智能机器人在获取了所述用户的所述声纹模型后,自动读取所述用户的面部影像图形,获得所述面部影像模型。
  7. 如权利要求1所述的快速识别方法,其特征在于,所述个人档案包括历史记录和收藏列表,所述家庭智能机器人接收已识别用户的指令,根据所述已识别用户的所述个人档案中的所述历史记录或所述收藏列表执行所述指令。
  8. 如权利要求1所述的快速识别方法,其特征在于,提供一存储单元用以存储复数个与时间关联的预录制语音,所述个人档案包括所述用户的姓名,所述家庭智能机器人对所述用户自动进行面部影像识别,根据识别结果调取所述用户的所述个人档案中的所述姓名,并根据当前时间于所述存储单元中选择对应的所述预录制语音,将所述姓名通过机器发声与所述预录制语音拼接后播放。
  9. 如权利要求1所述的快速识别方法,其特征在于,提供一摄像头对所述用户的面部影像图像进行读取。
  10. 一种家庭智能机器人,其特征在于,采用上述权利要求1-9中任一项所述的快速识别方法。
PCT/CN2016/101567 2015-10-09 2016-10-09 一种快速识别方法及家庭智能机器人 WO2017059815A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US15/766,890 US10664511B2 (en) 2015-10-09 2016-10-09 Fast identification method and household intelligent robot
JP2018517885A JP6620230B2 (ja) 2015-10-09 2016-10-09 迅速な識別方法及び家庭用の知能ロボット
EP16853119.2A EP3361410A4 (en) 2015-10-09 2016-10-09 FAST IDENTIFICATION PROCESS AND INTELLIGENT BUDGET ROBOT

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510650110.7A CN106570443A (zh) 2015-10-09 2015-10-09 一种快速识别方法及家庭智能机器人
CN201510650110.7 2015-10-09

Publications (1)

Publication Number Publication Date
WO2017059815A1 true WO2017059815A1 (zh) 2017-04-13

Family

ID=58487272

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/101567 WO2017059815A1 (zh) 2015-10-09 2016-10-09 一种快速识别方法及家庭智能机器人

Country Status (6)

Country Link
US (1) US10664511B2 (zh)
EP (1) EP3361410A4 (zh)
JP (1) JP6620230B2 (zh)
CN (1) CN106570443A (zh)
TW (1) TWI621470B (zh)
WO (1) WO2017059815A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107623614A (zh) * 2017-09-19 2018-01-23 百度在线网络技术(北京)有限公司 用于推送信息的方法和装置
CN109688183A (zh) * 2018-08-20 2019-04-26 深圳壹账通智能科技有限公司 群控设备识别方法、装置、设备及计算机可读存储介质
JPWO2018216180A1 (ja) * 2017-05-25 2019-11-07 三菱電機株式会社 音声認識装置および音声認識方法

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106570443A (zh) 2015-10-09 2017-04-19 芋头科技(杭州)有限公司 一种快速识别方法及家庭智能机器人
WO2019084963A1 (zh) * 2017-11-06 2019-05-09 深圳市沃特沃德股份有限公司 机器人及其服务方法和装置
WO2019176018A1 (ja) * 2018-03-14 2019-09-19 株式会社ウフル Aiスピーカーシステム、aiスピーカーシステムの制御方法、及びプログラム
WO2019181144A1 (ja) * 2018-03-20 2019-09-26 ソニー株式会社 情報処理装置及び情報処理方法、並びにロボット装置
CN108765921A (zh) * 2018-04-04 2018-11-06 昆山市工研院智能制造技术有限公司 基于视觉语意分析应用于巡逻机器人的智能巡逻方法
CN110390938A (zh) * 2018-04-20 2019-10-29 比亚迪股份有限公司 基于声纹的语音处理方法、装置和终端设备
CN110290846B (zh) * 2018-06-29 2021-09-24 深圳市大疆创新科技有限公司 一种虚拟对战的处理方法、服务器以及可移动平台
CN109446876B (zh) * 2018-08-31 2020-11-06 百度在线网络技术(北京)有限公司 手语信息处理方法、装置、电子设备和可读存储介质
CN111177329A (zh) * 2018-11-13 2020-05-19 奇酷互联网络科技(深圳)有限公司 一种智能终端的用户交互方法、智能终端及存储介质
CN111292734B (zh) * 2018-12-06 2024-03-08 阿里巴巴集团控股有限公司 一种语音交互方法和装置
CN109882985B (zh) * 2018-12-26 2020-07-28 珠海格力电器股份有限公司 一种语音播报方法、装置、存储介质及空调
US11151993B2 (en) * 2018-12-28 2021-10-19 Baidu Usa Llc Activating voice commands of a smart display device based on a vision-based mechanism

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060047362A1 (en) * 2002-12-02 2006-03-02 Kazumi Aoyama Dialogue control device and method, and robot device
CN103345232A (zh) * 2013-07-15 2013-10-09 孟凡忠 个性化智能家居控制方法及系统
CN104700018A (zh) * 2015-03-31 2015-06-10 江苏祥和电子科技有限公司 一种用于智能机器人的识别方法
CN204462847U (zh) * 2014-12-28 2015-07-08 青岛通产软件科技有限公司 一种多功能酒店服务机器人
CN104951077A (zh) * 2015-06-24 2015-09-30 百度在线网络技术(北京)有限公司 基于人工智能的人机交互方法、装置和终端设备

Family Cites Families (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6347261B1 (en) * 1999-08-04 2002-02-12 Yamaha Hatsudoki Kabushiki Kaisha User-machine interface system for enhanced interaction
EP1079615A3 (en) * 1999-08-26 2002-09-25 Matsushita Electric Industrial Co., Ltd. System for identifying and adapting a TV-user profile by means of speech technology
JP2001277166A (ja) * 2000-03-31 2001-10-09 Sony Corp ロボット及びロボットの行動決定方法
EP1395803B1 (en) * 2001-05-10 2006-08-02 Koninklijke Philips Electronics N.V. Background learning of speaker voices
JP2003122392A (ja) * 2001-10-16 2003-04-25 Yamatake Corp 音声入力者を判定する方法及び装置
JP3951235B2 (ja) * 2003-02-19 2007-08-01 ソニー株式会社 学習装置及び学習方法並びにロボット装置
JP2005157086A (ja) * 2003-11-27 2005-06-16 Matsushita Electric Ind Co Ltd 音声認識装置
JP4595436B2 (ja) * 2004-03-25 2010-12-08 日本電気株式会社 ロボット、その制御方法及び制御用プログラム
JP2008233345A (ja) * 2007-03-19 2008-10-02 Toshiba Corp インタフェース装置及びインタフェース処理方法
CN101925916B (zh) * 2007-11-21 2013-06-19 高通股份有限公司 基于媒体偏好控制电子设备的方法和系统
US20180144649A1 (en) * 2010-06-07 2018-05-24 Affectiva, Inc. Smart toy interaction using image analysis
US10176725B2 (en) * 2011-08-29 2019-01-08 Worcester Polytechnic Institute System and method of pervasive developmental disorder interventions
CN102447786A (zh) * 2011-11-14 2012-05-09 候万春 一种个人专用生活协助装置和方法
CN202736475U (zh) * 2011-12-08 2013-02-13 华南理工大学 一种聊天机器人
US9117449B2 (en) * 2012-04-26 2015-08-25 Nuance Communications, Inc. Embedded system for construction of small footprint speech recognition with user-definable constraints
EP2852881A4 (en) * 2012-05-22 2016-03-23 Intouch Technologies Inc GRAPHIC USER INTERFACES CONTAINING TOUCH PAD TOUCH INTERFACES FOR TELEMEDICINE DEVICES
US8706827B1 (en) * 2012-06-21 2014-04-22 Amazon Technologies, Inc. Customized speech generation
TW201408052A (zh) * 2012-08-14 2014-02-16 Kentec Inc 電視裝置及其虛擬主持人顯示方法
JP2014092777A (ja) * 2012-11-06 2014-05-19 Magic Hand:Kk モバイル通信機器の音声による起動
CN103973441B (zh) * 2013-01-29 2016-03-09 腾讯科技(深圳)有限公司 基于音视频的用户认证方法和装置
FR3011375B1 (fr) * 2013-10-01 2017-01-27 Aldebaran Robotics Procede de dialogue entre une machine, telle qu'un robot humanoide, et un interlocuteur humain, produit programme d'ordinateur et robot humanoide pour la mise en œuvre d'un tel procede
US9489171B2 (en) * 2014-03-04 2016-11-08 Microsoft Technology Licensing, Llc Voice-command suggestions based on user identity
JP2015176058A (ja) * 2014-03-17 2015-10-05 株式会社東芝 電子機器、方法及びプログラム
US9538005B1 (en) * 2014-09-19 2017-01-03 Amazon Technologies, Inc. Automated response system
KR102299764B1 (ko) * 2014-11-28 2021-09-09 삼성전자주식회사 전자장치, 서버 및 음성출력 방법
CN104834849B (zh) * 2015-04-14 2018-09-18 北京远鉴科技有限公司 基于声纹识别和人脸识别的双因素身份认证方法及系统
CN106570443A (zh) 2015-10-09 2017-04-19 芋头科技(杭州)有限公司 一种快速识别方法及家庭智能机器人
US10289076B2 (en) * 2016-11-15 2019-05-14 Roborus Co., Ltd. Concierge robot system, concierge service method, and concierge robot
US10120386B2 (en) * 2016-11-18 2018-11-06 Robert Bosch Start-Up Platform North America, LLC, Series 1 Robotic creature and method of operation
US20180187969A1 (en) * 2017-01-03 2018-07-05 Samsung Electronics Co., Ltd. Refrigerator

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060047362A1 (en) * 2002-12-02 2006-03-02 Kazumi Aoyama Dialogue control device and method, and robot device
CN103345232A (zh) * 2013-07-15 2013-10-09 孟凡忠 个性化智能家居控制方法及系统
CN204462847U (zh) * 2014-12-28 2015-07-08 青岛通产软件科技有限公司 一种多功能酒店服务机器人
CN104700018A (zh) * 2015-03-31 2015-06-10 江苏祥和电子科技有限公司 一种用于智能机器人的识别方法
CN104951077A (zh) * 2015-06-24 2015-09-30 百度在线网络技术(北京)有限公司 基于人工智能的人机交互方法、装置和终端设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3361410A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2018216180A1 (ja) * 2017-05-25 2019-11-07 三菱電機株式会社 音声認識装置および音声認識方法
CN107623614A (zh) * 2017-09-19 2018-01-23 百度在线网络技术(北京)有限公司 用于推送信息的方法和装置
CN109688183A (zh) * 2018-08-20 2019-04-26 深圳壹账通智能科技有限公司 群控设备识别方法、装置、设备及计算机可读存储介质

Also Published As

Publication number Publication date
CN106570443A (zh) 2017-04-19
EP3361410A4 (en) 2019-03-27
EP3361410A1 (en) 2018-08-15
US10664511B2 (en) 2020-05-26
JP2018533064A (ja) 2018-11-08
US20180293236A1 (en) 2018-10-11
TW201713402A (zh) 2017-04-16
JP6620230B2 (ja) 2019-12-11
TWI621470B (zh) 2018-04-21

Similar Documents

Publication Publication Date Title
WO2017059815A1 (zh) 一种快速识别方法及家庭智能机器人
US20210005202A1 (en) Focus Session at a Voice Interface Device
US20210280172A1 (en) Voice Response Method and Device, and Smart Device
WO2018000280A1 (zh) 基于多模态的智能机器人交互方法和智能机器人
KR20140089863A (ko) 디스플레이 장치, 및 이의 제어 방법, 그리고 음성 인식 시스템의 디스플레이 장치 제어 방법
US11457061B2 (en) Creating a cinematic storytelling experience using network-addressable devices
CN105204357A (zh) 智能家居设备的情景模式调整方法及装置
TW201606760A (zh) 從音頻訊號的即時情緒辨識
WO2017084185A1 (zh) 基于语义分析的智能终端控制方法、系统及智能终端
WO2019007249A1 (zh) 一种交互方法、电子设备及服务器
WO2017141530A1 (ja) 情報処理装置、情報処理方法、及びプログラム
CN108154878A (zh) 控制监控设备的方法及装置
CN104883503A (zh) 基于语音的个性化拍照技术
CN111508491A (zh) 一种基于深度学习的智能语音交互设备
WO2020079941A1 (ja) 情報処理装置及び情報処理方法、並びにコンピュータプログラム
TWI594857B (zh) 一種對機器人進行訓練的系統及方法
WO2020003851A1 (ja) 音声処理装置、音声処理方法及び記録媒体
WO2019176252A1 (ja) 情報処理装置、情報処理システム、および情報処理方法、並びにプログラム
JPWO2020003785A1 (ja) 音声処理装置、音声処理方法及び記録媒体
CN111506183A (zh) 一种智能终端及用户交互方法
WO2020202862A1 (ja) 応答生成装置及び応答生成方法
CN111739528A (zh) 一种交互方法、装置和耳机
KR102089593B1 (ko) 디스플레이 장치, 및 이의 제어 방법, 그리고 음성 인식 시스템의 디스플레이 장치 제어 방법
KR102124396B1 (ko) 디스플레이 장치, 및 이의 제어 방법, 그리고 음성 인식 시스템의 디스플레이 장치 제어 방법
WO2018023521A1 (zh) 一种语音控制音乐网络播放方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16853119

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2018517885

Country of ref document: JP

WWE Wipo information: entry into national phase

Ref document number: 15766890

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2016853119

Country of ref document: EP