WO2017000775A1 - 一种机器人语音寻向转动系统及方法 - Google Patents

一种机器人语音寻向转动系统及方法 Download PDF

Info

Publication number
WO2017000775A1
WO2017000775A1 PCT/CN2016/085761 CN2016085761W WO2017000775A1 WO 2017000775 A1 WO2017000775 A1 WO 2017000775A1 CN 2016085761 W CN2016085761 W CN 2016085761W WO 2017000775 A1 WO2017000775 A1 WO 2017000775A1
Authority
WO
WIPO (PCT)
Prior art keywords
voice
sound signal
signal
unit
robot
Prior art date
Application number
PCT/CN2016/085761
Other languages
English (en)
French (fr)
Inventor
刘鑫
高鹏
范利春
施家琪
蔡鹏�
蔡明峻
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 EP16817139.5A priority Critical patent/EP3318888A4/en
Priority to US15/739,330 priority patent/US10343287B2/en
Priority to JP2017567761A priority patent/JP6608466B2/ja
Publication of WO2017000775A1 publication Critical patent/WO2017000775A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S3/00—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received
    • G01S3/80—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received using ultrasonic, sonic or infrasonic waves
    • G01S3/801—Details
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S3/00—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received
    • G01S3/80—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received using ultrasonic, sonic or infrasonic waves
    • G01S3/802—Systems for determining direction or deviation from predetermined direction
    • G01S3/808—Systems for determining direction or deviation from predetermined direction using transducers spaced apart and measuring phase or time difference between signals therefrom, i.e. path-difference systems
    • G01S3/8083—Systems for determining direction or deviation from predetermined direction using transducers spaced apart and measuring phase or time difference between signals therefrom, i.e. path-difference systems determining direction of source
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/78—Detection of presence or absence of voice signals

Definitions

  • the present invention relates to the field of audio recognition, and in particular, to a robot voice homing rotation system and method.
  • Sound source localization technology is a technique that uses acoustic and electronic devices to receive target sound field information to determine the location of a target sound source.
  • the sound source localization technology in the robot system is limited to the positioning of the sound source with higher energy value, and is not aimed at the speech of the user, so it is easily interfered by the environmental noise during the sound source localization, and the robot is facing the direction. After judging the error, there will be a phenomenon of swinging left and right, and it will also affect the microphone array beam alignment with the user who is talking, reduce the signal-to-noise ratio of the microphone array pickup, and have a certain impact on the user experience of speech recognition.
  • an aim is to realize a direction of locating the user's speaking position to drive the robot to turn to the user's speaking direction, thereby improving the accuracy and signal-to-noise ratio of the speech recognition.
  • Robot voice homing rotation system and method
  • a robot voice homing rotation system comprising:
  • a voice collection unit for collecting and storing a sound signal
  • a voice activity detecting unit connected to the voice collecting unit, configured to detect whether a voice signal exists in the sound signal, and output a detection result
  • a voice finder rotating unit connected to the voice activity detecting unit, when the detection result is that the voice signal is present in the sound signal, controlling the voice collecting unit to acquire position information corresponding to the sound signal, and outputting The location information;
  • a voice finder unit which is respectively connected to the voice collection unit and the voice finder rotation unit, and configured to extract the sound signal corresponding to the location information according to the location information, according to a location corresponding to the location information
  • the sound signal is calculated to obtain a homing angle value, and the homing angle value is output.
  • the voice collection unit includes: an acquisition device, a recording module, and a storage module;
  • the collecting device is configured to collect the sound signal and output the sound signal
  • the recording module is connected to the collection device and the storage module, respectively, for storing the collected sound signal in the storage module.
  • the voice collection unit further includes:
  • a channel mixing module connected to the collecting device, configured to mix the left channel signal and the right channel signal in the sound signal to generate mixed sound information, and output the sound information.
  • the voice direction finding rotation unit comprises:
  • a receiving module configured to receive the detection result, the location information, and the homing angle value
  • the speech homing unit acquires the homing angle value
  • an execution module connected to the receiving module, configured to drive the robot to turn the main interface to the sound source direction of the sound signal according to the homing angle value.
  • the voice finder unit comprises:
  • an acquiring module configured to acquire, according to the location information, a left channel signal and a right channel signal in the sound signal corresponding to the location information in the voice collection unit.
  • the voice finder unit further includes:
  • the finder processing module is configured to connect the acquiring module to calculate the homing angle value of the sound source direction according to the left channel signal and the right channel signal in the sound signal.
  • the collecting device adopts an array microphone, and the array microphone is configured to convert the picked analog voice signal into a digital sound signal and output the sound signal.
  • a robot voice homing rotation method is applied to the robot voice homing rotation system, comprising the following steps:
  • step S2 detecting whether there is a voice signal in the sound signal, and if so, executing step S3, if not, executing step S1;
  • the voice activity detecting unit of the robot voice homing rotation system detects the received sound signal to determine whether there is a voice signal sent by the user, which can improve the accuracy of the voice recognition; and calculate the voice through the voice finder unit.
  • the homing angle of the signal uses the voice homing unit to drive the direction of the sound source of the robot to the sound signal according to the homing angle, thereby improving the accuracy and signal to noise ratio of the voice recognition.
  • the robot voice homing rotation method can accurately obtain the effective voice signal sent by the user, and improve the signal-to-noise ratio and the accuracy of voice recognition.
  • FIG. 1 is a block diagram of an embodiment of a robot voice homing rotation system according to the present invention
  • FIG. 2 is a flow chart of a method for a robot voice homing rotation method according to the present invention.
  • a robot voice homing rotation system includes:
  • a voice collecting unit 2 configured to collect and store a sound signal
  • the voice activity detecting unit 1 is connected to the voice collecting unit 2, configured to detect whether a voice signal exists in the sound signal, and output the detection result;
  • the voice finder rotation unit 4 is connected to the voice activity detecting unit 1.
  • the control voice collecting unit 2 acquires the position information corresponding to the sound signal, and outputs the position information;
  • the voice finder unit 3 is connected to the voice collecting unit 2 and the voice finder rotating unit 4, respectively, for extracting a sound signal corresponding to the position information according to the position information, and calculating according to the sound signal corresponding to the position information to obtain the homing angle value And output the homing angle value.
  • the sound signal may include external noise as well as a voice signal spoken by the user.
  • the voice activity detecting unit 1 detects the received sound signal, and when the voice signal exists in the sound signal, acquires the position information of the sound signal in the voice collecting unit 2, and sends the position information It is sent to the voice finder unit 3 for homing.
  • the voice activity detecting unit 1 detects the received voice signal to determine whether there is a voice signal sent by the user, so as to improve the accuracy of the voice recognition; and the voice finder unit 3 calculates the angle of the voice signal.
  • the voice direction finding rotation unit 4 drives the sound source direction of the robot to turn the sound signal according to the seek angle, thereby improving the accuracy and signal to noise ratio of the voice recognition.
  • the voice collection unit 2 includes: an acquisition device 21, a recording module 24, and a storage module 22;
  • the collecting device 21 is configured to collect a sound signal and output a sound signal
  • the recording module 24 is connected to the collection device 21 and the storage module 22 for storing the collected sound signals in the storage module 22.
  • the voice collection unit 2 further includes:
  • the channel mixing module 23 is connected to the collecting device 21 for mixing the left channel signal and the right channel signal in the sound signal to generate mixed sound information, and outputs the mixed sound information.
  • the left channel signal and the right channel signal output by the acquisition device 21 can be mixed and output by the channel mixing module 23.
  • the voice finder rotation unit 4 includes:
  • the receiving module 41 is configured to receive the detection result, the location information, and the homing angle value
  • control voice collecting unit 2 acquires the position information of the sound signal, and sends the position information to the voice finder unit 3, and obtains the homing angle value through the voice finder unit 3;
  • the execution module 42 is connected to the receiving module 41 for driving the robot to turn the main interface to the sound source direction of the standard voice signal according to the homing angle value.
  • the location information corresponding to the voice information is acquired and a notification event is sent, and the triggering event is triggered.
  • the sound finder unit 3 extracts the sound signal corresponding to the position information in the voice collecting unit 2 to calculate the acquisition homing angle value; when the voice activity detecting unit 1 outputs the detection result that the voice signal does not exist in the voice signal, the voice finder The unit 3 is in the standby state, and the voice activity detecting unit 1 is in the state of detecting the voice signal.
  • the voice finder unit 3 comprises:
  • the obtaining module 31 is configured to acquire, according to the location information, a left channel signal and a right channel signal in the sound signal corresponding to the location information in the voice collection unit 2.
  • the sound signal corresponding to the position information in the voice collecting unit 2 is extracted by the acquiring module 31, and an accurate homing angle value is obtained according to the sound signal, thereby improving the finder accuracy.
  • the voice finder unit 3 further includes:
  • the finder processing module 32 is connected to the acquisition module 31 for calculating the homing angle value of the sound source direction according to the left channel signal and the right channel signal in the sound signal.
  • the calculation is performed by the speech homing algorithm SO library in the homing processing module 32 to obtain the homing angle value of the sound source direction.
  • the acquisition device 21 employs an array microphone for converting the picked analog voice signal into a digital base voice signal and outputting it.
  • the array microphone in this embodiment is used to pick up sound signals, and by analyzing and processing the multi-path sound signals to obtain the position of the sound source, the execution module 42 can drive the array microphones to turn to the speaking user. Thereby improving the noise immunity of voice homing.
  • the robot voice homing rotation system realizes the positioning of the robot to the user's speaking orientation, and then rotates the robot body interface to align the microphone array beam with the speaking user.
  • the robot voice homing rotation system detects the sound signal by using the voice activity detecting unit 1 to determine whether there is a voice signal in the sound signal, and when there is a voice signal, the sound is sought
  • the rotating unit 4 sends a notification event, and after the voice finder rotating unit 4 listens to the notification event, the voice finder unit 3 is called first according to the voice activity detecting unit 1
  • the output speech position index is used to synchronously acquire the speech signals of the left and right channels of the array microphone, and then calculate and output the homing angle value to the speech homing rotation unit 4 through the speech homing algorithm SO library.
  • the voice homing rotation unit 4 calls the power system to turn the robot main interface to the user so that the microphone array beam is aimed at the user who is talking.
  • a robot voice homing rotation method is applied to a robot voice homing rotation system, which includes the following steps:
  • the robot voice homing rotation method realizes the positioning of the robot to the user's speaking orientation, and then rotates the robot body interface to align the microphone array beam with the speaking user.
  • the method can accurately obtain an effective voice signal sent by the user, and improve the signal-to-noise ratio and the accuracy of the voice recognition.

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Manipulator (AREA)
  • Circuit For Audible Band Transducer (AREA)
  • Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)

Abstract

一种机器人语音寻向转动系统及方法，机器人语音寻向转动系统采用的语音活动检测单元（1）对接收到的声音信号进行检测，判断是否存在用户发出的语音信号（S2）；通过语音寻向单元（3）计算语音信号的寻向角度，采用语音寻向转动单元（4）根据寻向角度驱动机器人转向声音信号的声源方向（S5）。采用机器人语音寻向转动方法可准确获取有效的用户发出的语音信号，提高信噪比及语音识别的准确率。

Description

一种机器人语音寻向转动系统及方法 技术领域
本发明涉及音频识别领域，尤其涉及一种机器人语音寻向转动系统及方法。
背景技术
声源定位技术是利用声学与电子装置接收目标声场信息以确定目标声源位置的一种技术。目前机器人系统中的声源定位技术还局限于对能量值较高声源的定位，并不是针对用户说话的语音进行定位，因此在声源定位的过程中容易受到环境噪声的干扰，机器人对方向判断出错后，会出现左右乱摆的现象，同时也会影响到麦克风阵列波束对准正在说话的用户，降低麦克风阵列拾音的信噪比，对语音识别的用户体验造成一定的影响。
发明内容
针对现有的声源定位技术存在的上述问题，现提供一种旨在实现可对用户说话方位进行寻向，以驱动机器人转向用户的说话方向，从而提高语音识别的准确率和信噪比的机器人语音寻向转动系统及方法。
具体技术方案如下：
一种机器人语音寻向转动系统，包括：
语音采集单元，用于采集并存储声音信号；
语音活动检测单元，连接所述语音采集单元，用于检测所述声音信号中是否存在语音信号，并输出检测结果；
语音寻向转动单元，连接所述语音活动检测单元，当所述检测结果为所述声音信号中存在所述语音信号时，控制所述语音采集单元获取所述声音信号对应的位置信息，并输出所述位置信息；
语音寻向单元，分别连接所述语音采集单元和所述语音寻向转动单元，用于根据所述位置信息提取与所述位置信息对应的所述声音信号，根据与所述位置信息对应的所述声音信号进行计算以获取寻向角度值，并输出所述寻向角度值。
优选的，所述语音采集单元包括：采集设备、录音模块和存储模块；
所述采集设备，用以采集所述声音信号，并输出所述声音信号；
所述录音模块分别连接所述采集设备和存储模块，用以将采集到的所述声音信号存储于所述存储模块中。
优选的，所述语音采集单元还包括：
声道混合模块，连接所述采集设备，用于将所述声音信号中的左声道信号和右声道信号混合生成混合声音信息，并输出。
优选的，所述语音寻向转动单元包括：
接收模块，用于接收所述检测结果、所述位置信息和所述寻向角度值；
当所述检测结果为所述声音信号中存在所述语音信号时，控制所述语音采集单元获取所述声音信号的位置信息，并将所述位置信息发送至所述语音寻向单元，通过所述语音寻向单元获取所述寻向角度值；
执行模块，连接所述接收模块，用于根据所述寻向角度值驱动机器人将主界面转向所述声音信号的声源方向。
优选的，所述语音寻向单元包括：
获取模块，用于根据所述位置信息获取所述语音采集单元中的与所述位置信息对应的所述声音信号中的左声道信号和右声道信号。
优选的，所述语音寻向单元还包括：
寻向处理模块，连接所述获取模块，用以根据所述声音信号中的左声道信号和右声道信号进行计算获取声源方向的所述寻向角度值。
优选的，所述采集设备采用阵列麦克风，所述阵列麦克风用以将拾取的模拟的语音信号转换为数字的所述声音信号并输出。
一种机器人语音寻向转动方法，应用于所述机器人语音寻向转动系统，包括下述步骤：
S1.获取并存储所述声音信号；
S2.检测所述声音信号中是否存在语音信号，若是，执行步骤S3，若否，执行步骤S1；
S3.获取所述声音信号的位置信息；
S4.根据所述位置信息提取与所述位置信息对应的所述声音信号，根据与所述位置信息对应的所述声音信号进行计算以获取所述寻向角度值；
S5.根据所述寻向角度值驱动机器人将主界面转向所述声音信号的声源方向。
上述技术方案的有益效果：
在本技术方案中，机器人语音寻向转动系统采用的语音活动检测单元对接收到的声音信号进行检测判断是否存在用户发出的语音信号，可提高语音识别的准确率；通过语音寻向单元计算语音信号的寻向角度，采用语音寻向转动单元根据寻向角度驱动机器人转向声音信号的声源方向，从而提高语音识别的准确率及信噪比。采用机器人语音寻向转动方法可准确获取有效的用户发出的语音信号，提高信噪比及语音识别的准确率。
附图说明
图1为本发明所述机器人语音寻向转动系统的一种实施例模块图；
图2为本发明所述机器人语音寻向转动方法的方法流程图。
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。
如图1所示，一种机器人语音寻向转动系统，包括：
语音采集单元2，用于采集并存储声音信号；
语音活动检测单元1，连接语音采集单元2，用于检测声音信号中是否存在语音信号，并输出检测结果；
语音寻向转动单元4，连接语音活动检测单元1，当检测结果为声音信号中存在语音信号时，控制语音采集单元2获取声音信号对应的位置信息，并输出位置信息；
语音寻向单元3，分别连接语音采集单元2和语音寻向转动单元4，用于根据位置信息提取与位置信息对应的声音信号，根据与位置信息对应的声音信号进行计算以获取寻向角度值，并输出寻向角度值。
进一步地，声音信号可包括外界的噪音以及用户说话的语音信号。采用语音活动检测单元1对接收到的声音信号进行检测，当声音信号中存在语音信号时，获取该声音信号在语音采集单元2中的位置信息，并将位置信息发 送至语音寻向单元3进行寻向。
在本实施例中，采用语音活动检测单元1对接收到的声音信号进行检测判断是否存在用户发出的语音信号，以提高语音识别的准确率；通过语音寻向单元3计算语音信号的寻向角度，采用语音寻向转动单元4根据寻向角度驱动机器人转向声音信号的声源方向，从而提高了语音识别的准确率及信噪比。本实施例中的语音活动检测单元1对语音信号检测的过程为现有技术，关于语音信号获得的具体过程此处不再赘述。
在优选的实施例中，语音采集单元2包括：采集设备21、录音模块24和存储模块22；
采集设备21，用以采集声音信号，并输出声音信号；
录音模块24分别连接采集设备21和存储模块22，用以将采集到的声音信号存储于存储模块22中。
在优选的实施例中，语音采集单元2还包括：
声道混合模块23，连接采集设备21，用于将声音信号中的左声道信号和右声道信号混合生成混合声音信息，并输出。
在本实施例中，采用声道混合模块23可将采集设备21输出的左声道信号和右声道信号混合并输出。
在优选的实施例中，语音寻向转动单元4包括：
接收模块41，用于接收检测结果、位置信息和寻向角度值；
当检测结果为声音信号中存在语音信号时，控制语音采集单元2获取声音信号的位置信息，并将位置信息发送至语音寻向单元3，通过语音寻向单元3获取寻向角度值；
执行模块42，连接接收模块41，用于根据寻向角度值驱动机器人将主界面转向标准语音信号的声源方向。
在本实施例中，当语音活动检测单元1输出的检测结果为声音信号中存在语音信号时，获取该声音信息对应的位置信息并发出通知事件，并触发语 音寻向单元3提取语音采集单元2中与该位置信息对应的声音信号，以计算获取寻向角度值；当语音活动检测单元1输出的检测结果为声音信号中不存在语音信号时，语音寻向单元3处于待机状态，语音活动检测单元1处于检测语音信号状态。
在优选的实施例中，语音寻向单元3包括：
获取模块31，用于根据位置信息获取语音采集单元2中的与位置信息对应的声音信号中的左声道信号和右声道信号。
在本实施例中，通过获取模块31提取语音采集单元2中与位置信息对应的声音信号，根据该声音信号获取准确的寻向角度值，从而提高寻向准确度。
在优选的实施例中，语音寻向单元3还包括：
寻向处理模块32，连接获取模块31，用以根据声音信号中的左声道信号和右声道信号进行计算获取声源方向的寻向角度值。
在本实施例中，通过寻向处理模块32中的语音寻向算法SO库进行计算以获取声源方向的寻向角度值。
在优选的实施例中，采集设备21采用阵列麦克风，阵列麦克风用以将拾取的模拟的语音信号转换为数字的基础语音信号并输出。
在本实施例中的阵列麦克风用于拾取声音信号，通过对多路声音信号进行分析与处理，以获得声源的位置，执行模块42可驱动阵列麦克风转向正在说话的用户。从而提高语音寻向的抗噪能力。
本实施例提供的机器人语音寻向转动系统实现机器人对用户说话方位的定位，随后转动机器人主体界面，将麦克风阵列波束对准正在说话的用户。机器人语音寻向转动系统在从语音采集单元2中采集声音信号后，采用语音活动检测单元1对声音信号进行检测，以判断声音信号中是否存在语音信号，当存在语音信号时，向语音寻向转动单元4发送通知事件，语音寻向转动单元4在监听到通知事件后，调用语音寻向单元3先根据语音活动检测单元1 输出的语音位置索引来同步获取阵列麦克风左右声道的语音信号，随后通过语音寻向算法SO库计算并输出寻向角度值至语音寻向转动单元4。语音寻向转动单元4在获取寻向角度值之后，调用动力系统将机器人主界面转向用户，使得麦克风阵列波束对准正在说话的用户。
如图2所示，一种机器人语音寻向转动方法，应用于机器人语音寻向转动系统，包括下述步骤：
S1.获取并存储声音信号；
S2.检测声音信号中是否存在语音信号，并输出检测结果，若是，执行步骤S3，若否，执行步骤S1；
S3.获取语音信号的位置信息；
S4.根据位置信息提取与位置信息对应的声音信号，根据与位置信息对应的声音信号进行计算以获取寻向角度值；
S5.根据寻向角度值驱动机器人将主界面转向标准语音信号的声源方向。
本实施例提供的机器人语音寻向转动方法实现机器人对用户说话方位的定位，随后转动机器人主体界面，将麦克风阵列波束对准正在说话的用户。该方法可准确获取有效的用户发出的语音信号，提高信噪比及语音识别的准确率。
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 (8)

  1. 一种机器人语音寻向转动系统，其特征在于，包括：
    语音采集单元，用于采集并存储声音信号；
    语音活动检测单元，连接所述语音采集单元，用于检测所述声音信号中是否存在语音信号，并输出检测结果；
    语音寻向转动单元，连接所述语音活动检测单元，当所述检测结果为所述声音信号中存在所述语音信号时，控制所述语音采集单元获取所述声音信号对应的位置信息，并输出所述位置信息；
    语音寻向单元，分别连接所述语音采集单元和所述语音寻向转动单元，用于根据所述位置信息提取与所述位置信息对应的所述声音信号，根据与所述位置信息对应的所述声音信号进行计算以获取寻向角度值，并输出所述寻向角度值。
  2. 如权利要求1所述机器人语音寻向转动系统，其特征在于，所述语音采集单元包括：采集设备、录音模块和存储模块；
    所述采集设备，用以采集所述声音信号，并输出所述声音信号；
    所述录音模块分别连接所述采集设备和存储模块，用以将采集到的所述声音信号存储于所述存储模块中。
  3. 如权利要求2所述机器人语音寻向转动系统，其特征在于，所述语音采集单元还包括：
    声道混合模块，连接所述采集设备，用于将所述声音信号中的左声道信号和右声道信号混合生成混合声音信息，并输出。
  4. 如权利要求1所述机器人语音寻向转动系统，其特征在于，所述语音寻向转动单元包括：
    接收模块，用于接收所述检测结果、所述位置信息和所述寻向角度值；
    当所述检测结果为所述声音信号中存在所述语音信号时，控制所述语音 采集单元获取所述声音信号的位置信息，并将所述位置信息发送至所述语音寻向单元，通过所述语音寻向单元获取所述寻向角度值；
    执行模块，连接所述接收模块，用于根据所述寻向角度值驱动机器人将主界面转向所述声音信号的声源方向。
  5. 如权利要求1所述机器人语音寻向转动系统，其特征在于，所述语音寻向单元包括：
    获取模块，用于根据所述位置信息获取所述语音采集单元中的与所述位置信息对应的所述声音信号中的左声道信号和右声道信号。
  6. 如权利要求5所述机器人语音寻向转动系统，其特征在于，所述语音寻向单元还包括：
    寻向处理模块，连接所述获取模块，用以根据所述声音信号中的左声道信号和右声道信号进行计算获取声源方向的所述寻向角度值。
  7. 如权利要求2所述机器人语音寻向转动系统，其特征在于，所述采集设备采用阵列麦克风，所述阵列麦克风用以将拾取的模拟的语音信号转换为数字的所述声音信号并输出。
  8. 一种机器人语音寻向转动方法，应用于权利要求1所述机器人语音寻向转动系统，其特征在于，包括下述步骤：
    S1.获取并存储所述声音信号；
    S2.检测所述声音信号中是否存在语音信号，若是，执行步骤S3，若否，执行步骤S1；
    S3.获取所述声音信号的位置信息；
    S4.根据所述位置信息提取与所述位置信息对应的所述声音信号，根据与所述位置信息对应的所述声音信号进行计算以获取所述寻向角度值；
    S5.根据所述寻向角度值驱动机器人将主界面转向所述声音信号的声源方向。
PCT/CN2016/085761 2015-06-30 2016-06-14 一种机器人语音寻向转动系统及方法 WO2017000775A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP16817139.5A EP3318888A4 (en) 2015-06-30 2016-06-14 SYSTEM AND METHOD FOR ROBOT ROTATION BASED ON VOICE DIRECTION SEARCH
US15/739,330 US10343287B2 (en) 2015-06-30 2016-06-14 Robot voice direction-seeking turning system and method
JP2017567761A JP6608466B2 (ja) 2015-06-30 2016-06-14 ロボット音声探索回転システム及び方法

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510388219.8 2015-06-30
CN201510388219.8A CN106328130A (zh) 2015-06-30 2015-06-30 一种机器人语音寻向转动系统及方法

Publications (1)

Publication Number Publication Date
WO2017000775A1 true WO2017000775A1 (zh) 2017-01-05

Family

ID=57607869

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/085761 WO2017000775A1 (zh) 2015-06-30 2016-06-14 一种机器人语音寻向转动系统及方法

Country Status (6)

Country Link
EP (1) EP3318888A4 (zh)
JP (1) JP6608466B2 (zh)
CN (1) CN106328130A (zh)
HK (1) HK1231623A1 (zh)
TW (1) TW201719632A (zh)
WO (1) WO2017000775A1 (zh)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106847298B (zh) * 2017-02-24 2020-07-21 海信集团有限公司 一种基于弥漫式语音交互的拾音方法和装置
CN107358950A (zh) * 2017-05-27 2017-11-17 芜湖星途机器人科技有限公司 服务机器人语音交互系统
CN107297745B (zh) * 2017-06-28 2019-08-13 上海木木机器人技术有限公司 语音交互方法、语音交互装置及机器人
CN109688512B (zh) * 2018-12-26 2020-12-22 北京猎户星空科技有限公司 一种拾音方法及装置
CN109920425B (zh) * 2019-04-03 2021-11-12 北京石头世纪科技股份有限公司 机器人语音控制方法、装置、机器人和介质
CN110428722A (zh) * 2019-08-07 2019-11-08 杭州任你说智能科技有限公司 一种基于智能语音交互技术的智能地球仪及其工作方法
TWI736117B (zh) * 2020-01-22 2021-08-11 瑞昱半導體股份有限公司 聲音定位裝置與方法
CN113464858B (zh) * 2021-08-03 2023-02-28 浙江欧菲克斯交通科技有限公司 移动式应急照明控制方法及装置
CN114242072A (zh) * 2021-12-21 2022-03-25 上海帝图信息科技有限公司 一种用于智能机器人的语音识别系统
CN116189679B (zh) * 2023-05-04 2023-09-26 北京水晶石数字科技股份有限公司 一种具有环境音降噪的语音交互装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100034397A1 (en) * 2006-05-10 2010-02-11 Honda Motor Co., Ltd. Sound source tracking system, method and robot
CN102305925A (zh) * 2011-07-22 2012-01-04 北京大学 一种机器人连续声源定位方法
CN103064061A (zh) * 2013-01-05 2013-04-24 河北工业大学 三维空间声源定位方法
CN103308889A (zh) * 2013-05-13 2013-09-18 辽宁工业大学 复杂环境下被动声源二维doa估计方法
CN104459625A (zh) * 2014-12-14 2015-03-25 南京理工大学 基于轨道移动双麦克风阵列的声源定位装置及方法

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3572849B2 (ja) * 1997-02-14 2004-10-06 富士ゼロックス株式会社 音源位置計測装置、及びカメラ撮影制御装置
US7215785B1 (en) * 2000-02-03 2007-05-08 Sang Gyu Ju Passive sound telemetry system and method and operating toy using the same
JP4815661B2 (ja) * 2000-08-24 2011-11-16 ソニー株式会社 信号処理装置及び信号処理方法
JP3771812B2 (ja) * 2001-05-28 2006-04-26 インターナショナル・ビジネス・マシーンズ・コーポレーション ロボットおよびその制御方法
JP2007221300A (ja) * 2006-02-15 2007-08-30 Fujitsu Ltd ロボット及びロボットの制御方法
WO2007138503A1 (en) * 2006-05-31 2007-12-06 Philips Intellectual Property & Standards Gmbh Method of driving a speech recognition system
JP2010197727A (ja) * 2009-02-25 2010-09-09 Nec Corp 音声認識装置、ロボット、音声認識方法、プログラム及び記録媒体
CN101763120B (zh) * 2009-12-17 2011-12-21 长安大学 一种基于声音引导的机器人定位装置及其定位方法
CN102411138A (zh) * 2011-07-13 2012-04-11 北京大学 一种机器人声源定位方法
CN103052001B (zh) * 2011-10-17 2015-06-24 联想(北京)有限公司 智能设备及其控制方法
CN102546953A (zh) * 2012-02-07 2012-07-04 深圳市金立通信设备有限公司 一种全语音操控移动终端的系统及方法
CN103366756A (zh) * 2012-03-28 2013-10-23 联想(北京)有限公司 一种声音信号的接收方法及装置
CN202622810U (zh) * 2012-05-15 2012-12-26 郑州大学 一种智能机器人的头颈部设计
CN103399293A (zh) * 2013-08-06 2013-11-20 成都四威高科技产业园有限公司 一种实现声源测向的方法及装置
CN104035065A (zh) * 2014-06-23 2014-09-10 河北工业大学 基于主动旋转的声源定向装置及其应用方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100034397A1 (en) * 2006-05-10 2010-02-11 Honda Motor Co., Ltd. Sound source tracking system, method and robot
CN102305925A (zh) * 2011-07-22 2012-01-04 北京大学 一种机器人连续声源定位方法
CN103064061A (zh) * 2013-01-05 2013-04-24 河北工业大学 三维空间声源定位方法
CN103308889A (zh) * 2013-05-13 2013-09-18 辽宁工业大学 复杂环境下被动声源二维doa估计方法
CN104459625A (zh) * 2014-12-14 2015-03-25 南京理工大学 基于轨道移动双麦克风阵列的声源定位装置及方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3318888A4 *

Also Published As

Publication number Publication date
HK1231623A1 (zh) 2017-12-22
JP6608466B2 (ja) 2019-11-20
JP2018526622A (ja) 2018-09-13
EP3318888A4 (en) 2019-03-27
TW201719632A (zh) 2017-06-01
EP3318888A1 (en) 2018-05-09
CN106328130A (zh) 2017-01-11

Similar Documents

Publication Publication Date Title
WO2017000775A1 (zh) 一种机器人语音寻向转动系统及方法
TWI667926B (zh) 一種基於麥克風陣列的拾音方法及系統
JP4204541B2 (ja) 対話型ロボット、対話型ロボットの音声認識方法および対話型ロボットの音声認識プログラム
TWI711035B (zh) 方位角估計的方法、設備、語音交互系統及儲存介質
JP6520878B2 (ja) 音声取得システムおよび音声取得方法
US20180186009A1 (en) Robot voice direction-seeking turning system and method
US9621984B1 (en) Methods to process direction data of an audio input device using azimuth values
WO2019080639A1 (zh) 一种对象识别方法、计算机设备及计算机可读存储介质
CN110875060A (zh) 语音信号处理方法、装置、系统、设备和存储介质
WO2015154445A1 (zh) 一种语音识别的方法及移动终端
KR100758707B1 (ko) 로봇 환경에서 음원 추적 시스템 및 방법
WO2014146547A1 (zh) 声音信号处理方法及设备
JP6467736B2 (ja) 音源位置推定装置、音源位置推定方法および音源位置推定プログラム
CN103278801A (zh) 一种变电站噪声成像侦测装置及侦测计算方法
CN103426440A (zh) 利用能量谱熵空间信息的语音端点检测装置及其检测方法
WO2020043037A1 (zh) 语音转录设备、系统、方法、及电子设备
JP7194897B2 (ja) 信号処理装置及び信号処理方法
KR20200129934A (ko) 오디오-비주얼 데이터에 기반한 화자 분리 방법 및 장치
CN112925235A (zh) 交互时的声源定位方法、设备和计算机可读存储介质
WO2019200722A1 (zh) 声源方向估计方法和装置
WO2016078415A1 (zh) 一种终端拾音控制方法、终端及终端拾音控制系统
CN203148562U (zh) 一种变电站噪声成像侦测装置
KR101791907B1 (ko) 위치 기반의 음향 처리 장치 및 방법
Martinson et al. Robotic discovery of the auditory scene
WO2013132216A1 (en) Method and apparatus for determining the number of sound sources in a targeted space

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: 16817139

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2017567761

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2016817139

Country of ref document: EP