WO2010020138A1 - Control method and device for monitoring equipment - Google Patents

Control method and device for monitoring equipment Download PDF

Info

Publication number
WO2010020138A1
WO2010020138A1 PCT/CN2009/072503 CN2009072503W WO2010020138A1 WO 2010020138 A1 WO2010020138 A1 WO 2010020138A1 CN 2009072503 W CN2009072503 W CN 2009072503W WO 2010020138 A1 WO2010020138 A1 WO 2010020138A1
Authority
WO
WIPO (PCT)
Prior art keywords
voice
speech recognition
operation instruction
sequence
recognition parameter
Prior art date
Application number
PCT/CN2009/072503
Other languages
French (fr)
Chinese (zh)
Inventor
崔志伟
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2010020138A1 publication Critical patent/WO2010020138A1/en

Links

Classifications

    • 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
    • G10L15/00—Speech recognition
    • G10L15/26—Speech to text systems

Definitions

  • the present invention relates to the field of communications, and in particular, to a method and an apparatus for controlling a monitoring device.
  • BACKGROUND With the development of network technologies and the improvement of user security awareness, the use of video surveillance for security defense has been widely used.
  • the pan/tilt is usually used to control the camera's monitoring angle.
  • the pan/tilt is a camera mounting platform consisting of two AC motors. The operator operates the pan/tilt to make the pan/tilt move horizontally or vertically. The camera realizes the change of the monitoring angle and performs video monitoring.
  • the implementation method of controlling the pan/tilt is:
  • the monitoring device client sends a control command to the monitoring device by using a mouse or a keyboard to control the monitoring device.
  • This monitoring device control method is complicated to operate, and only a professional can operate the monitoring device, and the method is not image and intuitive for the operator.
  • SUMMARY OF THE INVENTION The present invention has been made in view of the problem of complicated operation of a monitoring device control method existing in the related art. Therefore, it is a primary object of the present invention to provide a control method and apparatus for a monitoring device to solve the above problems. According to an aspect of the invention, a method of controlling a monitoring device is provided.
  • the control method of the monitoring device comprises: performing feature extraction on a digital voice signal corresponding to the received voice control instruction, acquiring a voice recognition parameter sequence of the voice control instruction; and pre-setting one or more voice recognition parameter sequences The search and the acquired speech recognition parameter sequence match the best speech recognition parameter sequence; determine the operation instruction corresponding to the searched speech recognition parameter sequence, and use the operation instruction to control the monitoring device.
  • the method further includes: receiving one or more voice control instructions in advance; performing feature extraction on the corresponding digital voice signal for each voice control instruction, acquiring and saving each voice control Instruction speech recognition parameter sequence; configuration speech recognition The correspondence between the parameter sequence and the operation instruction.
  • the above method further includes: saving, for each voice control instruction received in advance, a corresponding digital voice signal.
  • the operation for saving the corresponding digital voice signal for each voice control command received in advance is specifically: compressing the corresponding digital voice signal for each voice control signaling received in advance, and saving the compressed digital number voice signal.
  • the number of preset voice control instructions may be one or more.
  • the operation of controlling the device by using the operation instruction is specifically: sending an operation instruction to the target encoder, and the target encoder controls the monitoring device according to the operation instruction.
  • the operation of controlling the device by using the operation instruction is specifically: sending an operation instruction to the relay server, and the relay server forwards the operation instruction to the target encoder, and the target encoder controls the monitoring device according to the operation instruction.
  • the digital speech signal characteristic parameter corresponding to the voice control instruction may be extracted by using one of the following methods: a formant extraction method, an endpoint detection extraction method, a linear prediction to a general coefficient extraction method, a Mel cepstral coefficient extraction method, and a linear frequency extraction method.
  • the manner of finding the best speech recognition parameter sequence matching the acquired speech recognition parameter sequence comprises at least one of the following: a dynamic time rounding algorithm, a hidden Markov model.
  • a control device for a monitoring device is provided.
  • the control device of the monitoring device includes: an obtaining module, configured to perform feature extraction on a digital voice signal corresponding to the received voice control instruction, to obtain a voice recognition parameter sequence of the voice control instruction; and a matching module, configured to be preset The one or more speech recognition parameter sequences are matched with the acquired speech recognition parameter sequence to match the best speech recognition parameter sequence; the control module is configured to determine an operation instruction corresponding to the searched speech recognition parameter sequence, and monitor the operation instruction The device is controlled.
  • the device further includes: a receiving module, configured to receive one or more voice control commands in advance; and a saving module, configured to perform feature extraction on the digital voice signal corresponding to each voice control instruction, acquire and save each voice control Instruction speech recognition parameter sequence; configuration module, for Configure the correspondence between the speech recognition parameter sequence and the operation instruction.
  • the saving module is further configured to save the digital voice signal received in advance; or save the digital voice signal corresponding to each compressed voice control instruction.
  • FIG. 1 is a flowchart of a control method of a monitoring device according to an embodiment of a method of the present invention
  • FIG. 2 is a detailed processing flowchart of a control method of a monitoring device according to an embodiment of a method of the present invention
  • FIG. 4 is a structural block diagram of a control device of a monitoring device according to an embodiment of the device of the present invention
  • FIG. 5 is a specific embodiment of a control device for a monitoring device according to an embodiment of the device of the present invention
  • Structure frame diagram is intended to provide a further understanding of the invention, and are intended to be a part of the description of the invention.
  • FIG. 1 is a flowchart of a control method of a monitoring device according to an embodiment of a method of the present invention
  • FIG. 2 is a detailed processing flowchart of a control method of a monitoring device according to an embodiment of a method of the present invention
  • FIG. 4 is a structural block diagram of a control device of a monitoring device according to an embodiment
  • the present invention addresses a problem, and proposes a control scheme for the monitoring device, which uses a voice control command. Controlling the monitoring equipment, the scheme is simple to operate, and is more visual and intuitive than the prior art, and with the rapid development of technology, the speech recognition technology has gradually become a key technology of human-machine interface in information technology.
  • a control method of a monitoring device is provided.
  • the correspondence between the speech recognition parameter sequence and the operation instruction needs to be pre-configured.
  • one or more voice control commands may be received in advance, and the analog voice signal is converted for each voice control instruction.
  • a speech recognition parameter sequence of each speech control instruction is obtained, wherein there are various ways to extract the feature of the digital speech signal, for example, the formant extraction method can be used.
  • the obtained speech recognition parameter sequence may be saved in the form of a file on the hard disk or the memory of the computer.
  • the digital voice signal corresponding to each voice control instruction may be saved in the hard disk or the memory of the computer.
  • the digital voice signal can be compressed and then saved.
  • Step S102 Perform feature extraction on a digital voice signal corresponding to a received voice control command, and obtain the a voice recognition parameter sequence of the voice control command;
  • Step S104 searching for a sequence of the voice recognition parameter that matches the obtained voice recognition parameter sequence in the preset one or more voice recognition parameter sequences, for example, using dynamic time return The whole algorithm, the implicit Markov model, is matched.
  • Step S106 determining an operation instruction corresponding to the searched speech recognition parameter sequence, and controlling the monitoring device according to the determined operation instruction. Specifically, the operation instruction may be sent to the target encoder, and then the target encoder is configured according to the operation instruction. The monitoring device is controlled.
  • the operation command may be sent to the target encoder through the relay server.
  • the client sends an operation instruction to the relay server, and the relay server forwards the operation instruction to the target encoder, and the target encoder uses the The operating instructions control the monitoring device.
  • the monitoring device is controlled by using a voice instruction, so that the operator can control the monitoring device by directly inputting the voice control command, and the method is simple and more convenient for the operator.
  • Image, intuitive. 2 is a detailed processing flowchart of a method for controlling a monitoring device according to an embodiment of the method of the present invention. As shown in FIG.
  • Step S201 Different operating instructions are required for different monitoring devices. Corresponding voice control commands are preset according to different operation commands, for example, voice control commands for setting the pan/tilt to rotate leftward, rightward, upward, and downward.
  • Step S202 the operation instruction for each voice control commands, Bian speech signal samples, particular, preclude the sampling frequency of the speech signal may be 8KHZ, the 8KHZ suitable for simple statement identifying another 1 J, for complex statements, You can choose a higher sampling frequency, the sample interval is 10MS, the sample interval contains 80 sample points, a segment of speech contains multiple samples, and the short-term energy sum is calculated every 10MS, and the short-term energy obtained by the sample is greater than At a certain threshold, it is considered that the voice sample has begun. When the short-term energy is less than 1/20 of the average energy, the voice sample is considered to have ended, the voice sample signal is obtained, and the pulse-code modulation is utilized.
  • Step S203 extracting corresponding feature parameters for each of the voice sample signals obtained in step S202, and determining a sequence of voice recognition parameters, which may use a formant extraction method, an endpoint detection extraction method, a linear prediction to a general coefficient extraction method,
  • the feature parameters are extracted by the MFCC extraction method, the LSF extraction method, the PLCC extraction method, and the EPOCH extraction method.
  • the linear prediction coding (LPC) algorithm is taken as an example for description.
  • the obtained voice recognition parameter sequence and the voice PCM code obtained in step S202 are saved in the form of a file on a computer hard disk or a memory of the client, where the file name The name should correspond to the control instruction. For example, if the operation instruction is to rotate the monitoring device to the right, the file name can be Template_Right.
  • the saved file can be read into the hard disk or the memory at one time without having to repeatedly read in each time the voice command is recognized, thereby saving the time of the recognition process.
  • Step S205 the voice control command is input, and the voice control command is sampled.
  • the implementation method is the same as step S202, and details are not described herein again.
  • Step S207 matching the speech recognition parameter sequence S obtained in step S206 with a plurality of preset speech recognition parameter sequences, selecting the best matching speech recognition parameter, and according to the operation instruction corresponding to the best matching speech recognition parameter Control the monitoring equipment.
  • the input speech recognition parameter sequence S is sequentially matched with the reference templates A, B, C, and D stored in the template library, wherein the reference template with the highest matching degree is the recognition result, and the control instruction represented by the recognition result is determined according to the recognition result. If the speech recognition parameter sequence S matches the reference template A most, it is determined that the operation instruction executed on the monitoring device is the leftward (Template_Left) corresponding to the reference template A.
  • Step S208 the client establishes a TCP short connection with the decoder, and sends the monitoring device control request (ie, the operation instruction described above) determined in step S207 to the remote encoder, for example, the monitoring device control may be sent in an XML format.
  • the request, the monitoring device control request message may include information such as the ID of the monitoring device of the message destination, the manner of control, the direction of the control, and the amount of control step, and obtains the response message of the encoder to complete the control operation of the monitoring device.
  • FIG. 3 shows a specific implementation environment for implementing the method. As shown in FIG. 3, the client may send a monitoring device control message to the target encoder through the network, and the encoder provides a device for implementing streaming media data in response to the monitoring device.
  • the control request is directly connected to the monitoring device and the camera. After receiving the control device control request from the client, the 485 port sends a corresponding control command to the monitoring device to complete the control operation of the monitoring device.
  • the TCP short-chain connection mode is used, but the method is not limited thereto, and the client may negotiate according to the encoder.
  • TCP TCP long-chain connection or other connection methods such as UPD, and monitoring device control requests can also be implemented in formats other than XML.
  • the embodiment of the present invention uses a voice command to control the monitoring device to rotate left, right, up, and down, but is not limited to the second, and the technical solution provided by the present invention can support other devices for controlling the monitoring device and the camera. More operations, such as controlling camera zoom, adjusting brightness, and manipulating the monitoring equipment's accessories such as lights and wipers.
  • a computer readable medium having stored thereon computer executable instructions for causing a computer or processor to perform, for example, when executed by a computer or processor The processing of all the steps shown in Figures 1 and 2.
  • Apparatus Embodiment According to an embodiment of the present invention, a control apparatus for a monitoring apparatus is provided.
  • the device includes an acquisition module 10, a matching module 20, and a control module 30.
  • the above modules are described in detail below.
  • the obtaining module 10 is configured to perform feature extraction on the digital voice signal corresponding to the received voice control instruction, and obtain a voice recognition parameter sequence of the voice control instruction.
  • the matching module 20 is configured to search for a sequence of the best speech recognition parameter that matches the acquired sequence of the speech recognition parameter in the preset one or more speech recognition parameter sequences, and the module may be connected to the acquisition module 10.
  • the control module 30 is configured to determine an operation instruction corresponding to the searched speech recognition parameter sequence, and control the monitoring device by using an operation instruction, and the module may be connected to the matching module 20.
  • the control device of the monitoring device provided by the embodiment of the present invention controls the monitoring device by using a voice instruction, so that the operator can control the monitoring device by directly inputting the voice control command, and the operation is simple for the operator. , and more visual and intuitive.
  • 5 is a detailed structural diagram of a control device of a monitoring device according to an embodiment of the present invention, Based on the apparatus shown in FIG. 4, the apparatus shown in FIG. 5 further includes a receiving module 40, a saving module 50, and a configuration module 60. The above modules are described in detail.
  • the receiving module 40 is configured to receive one or more voice control commands in advance.
  • the saving module 50 is configured to perform feature extraction on the digital voice signal corresponding to each voice control instruction, acquire and save a voice recognition parameter sequence of each voice control instruction, and the module may be connected to the matching module 20 and the receiving module 40.
  • the configuration module 60 is configured to configure a correspondence between the speech recognition parameter sequence and the operation instruction, and the module may be connected to the saving module 50.
  • the saving module 50 is further configured to save the digital voice signal received in advance; or save the digital voice signal corresponding to each compressed voice control instruction.
  • the monitoring device is controlled by using a voice instruction, so that the operator can control the monitoring device by directly inputting the voice control command, for the operator
  • the method is simple, and more visual and intuitive.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Selective Calling Equipment (AREA)
  • Testing And Monitoring For Control Systems (AREA)
  • Telephonic Communication Services (AREA)

Abstract

A control method and device for a monitoring equipment, the method including: extracting the features of the digital speech signals corresponding to the received speech control instructions, obtaining the speech recognition parameter sequences of the speech control instructions; searching the speech recognition parameter sequences which are an optimal match with the obtained speech recognition parameter sequences in one or more pre-set speech recognition parameter sequences; determining the operating instructions corresponding to the searched speech recognition parameter sequences, and the operational instructions are utilized to control a monitoring apparatus.

Description

监控设备的控制方法和装置  Monitoring device control method and device
技术领域 本发明涉及通信领域， 尤其涉及一种监控设备的控制方法和装置。 背景技术 随着网络技术的发展以及用户安全防范意识的提高，利用视频监控进行 安全防范已经得到了较为普遍的应用。 在视频监控中， 通常使用云台来控制 摄像头的监控角度， 云台是由两个交流电机组成的摄像头安装平台， 操作人 员通过对云台进行操作， 使云台做水平或垂直的运动， 带动摄像头实现监控 角度的变化， 进行视频监控。 目前， 人们对云台进行控制的实现方法是： 监控装置客户端通过使用鼠 标或键盘， 向监控设备发送控制命令， 以控制监控设备。 这种监控设备控制 方法操作复杂， 只有专业人员才能对监控设备进行操作， 而且对于操作人员 来说， 该方法不够形象、 直观。 发明内容 考虑到相关技术中存在的监控设备控制方法操作复杂的问题而提出本 发明， 为此， 本发明的主要目的在于提供一种监控设备的控制方法及装置， 以解决上述问题。 根据本发明的一个方面， 提供了一种监控设备的控制方法。 才艮据本发明的监控设备的控制方法包括：对接收的语音控制指令对应的 数字语音信号进行特征提取， 获取语音控制指令的语音识别参数序列； 在预 先设置的一个或多个语音识别参数序列中查找与获取的语音识别参数序列匹 配最佳的语音识别参数序列； 确定查找的语音识别参数序列所对应的操作指 令， 并利用操作指令对监控设备进行控制。 其中， 在接收到语音控制指令之前， 该方法还包括： 预先接收一个或多 个语音控制指令； 对于每个语音控制指令， 将其对应的数字语音信号进行特 征提取， 获取并保存每个语音控制指令的语音识别参数序列； 配置语音识别 参数序列与操作指令之间的对应关系。 进一步地， 上述方法还包括： 对于预先接收的每个语音控制指令， 保存 其对应的数字语音信号。 其中， 上述对于预先接收的每个语音控制指令， 保存其对应的数字语音 信号的操作具体为： 对于预先接收的每个语音控制信令， 压缩其对应的数字 语音信号， 并保存压缩后的数字语音信号。 优选地， 对于每个操作指令， 预先设置的语音控制指令的数量可以为一 个或多个。 其中， 上述利用操作指令对设备进行控制的操作具体为： 向目标编码器 发送操作指令， 目标编码器艮据操作指令对监控设备进行控制。 其中， 上述利用操作指令对设备进行控制的操作具体为： 向中转服务器 发送操作指令， 中转服务器将操作指令转发给目标编码器， 目标编码器根据 操作指令对监控设备进行控制。 优选地，可以利用以下方式之一提取语音控制指令对应的数字语音信号 特征参数： 共振峰提取法、 端点检测提取法、 线性预测到普系数提取法、 Mel 倒谱系数提取法、 线性频率提取法。 优选地，查找与获取的语音识别参数序列匹配最佳的语音识别参数序列 的方式包括以下至少之一： 动态时间归整算法、 隐含马可夫模型。 根据本发明的另一个发面， 提供一种监控设备的控制装置。 根据本发明的监控设备的控制装置包括： 获取模块， 用于对接收的语音 控制指令对应的数字语音信号进行特征提取， 获取语音控制指令的语音识别 参数序列； 匹配模块， 用于在预先设置的一个或多个语音识别参数序列中查 找与获取的语音识别参数序列匹配最佳的语音识别参数序列； 控制模块， 用 于确定查找的语音识别参数序列所对应的操作指令， 并利用操作指令对监控 设备进行控制。 进一步地， 该装置还包括： 接收模块， 用于预先接收一个或多个语音控 制指令； 保存模块， 用于对每个语音控制指令对应的数字语音信号进行特征 提取， 获取并保存每个语音控制指令的语音识别参数序列； 配置模块， 用于 配置语音识别参数序列与操作指令之间的对应关系。 其中， 保存模块还用于保存预先接收的数字语音信号； 或者， 保存压缩 后的每个语音控制指令对应的数字语音信号。 通过本发明的上述至少一个技术方案，利用语音指令对监控设备进行控 制， 操作人员可以通过直接输入语音控制命令的方式对监控设备进行控制， 对于操作人员来说， 该方法操作简单， 且更加形象、 直观。 附图说明 附图用来提供对本发明的进一步理解， 并且构成说明书的一部分， 与本 发明的实施例一起用于解释本发明， 并不构成对本发明的限制。 在附图中： 图 1是根据本发明方法实施例的监控设备的控制方法的流程图； 图 2 是根据本发明方法实施例的监控设备的控制方法的详细处理流程 图； 图 3是根据本发明实施例的监控设备控制应用的环境部署图； 图 4是根据本发明装置实施例的监控设备的控制装置的结构框架图； 图 5 是根据本发明装置实施例的监控设备的控制装置的具体结构框架 图。 具体实施方式 功能相无述 如上所述， 目前， 人们在对监控设备进行控制时， 存在操作较为复杂的 问题， 本发明针对该问题， 提出一种监控设备的控制方案， 该方案利用语音 控制指令对监控设备进行控制， 该方案操作简单， 与现有技术相比， 更为形 象和直观， 而且随着科技的快速发展， 语音识别技术逐步成为信息技术中人 机接口的关键技术。 下面将结合附图详细描述本发明。 需要说明的是， 在不冲突的情况下， 本申请中的实施例及实施例中的特征可以相互组合。 方法实施例 才艮据本发明实施例， 提供了一种监控设备的控制方法。 在进行本发明实施例时，需要预先配置语音识别参数序列与操作指令之 间的对应关系， 具体地， 可以预先接收一个或多个语音控制指令， 对于每个 语音控制指令， 将模拟语音信号转换为数字语音信号并对上述数字语音信号 进行特征提取， 获取每个语音控制指令的语音识别参数序列， 其中， 对数字 语音信号进行特征提取的方式有多种， 例如， 可以釆用共振峰提取法、 端点 检测提取法、 线性预测到普系数提取法、 Mel 倒谱系数 （MFCC )提取法、 线性频率 （LSF ) 提取法、 PLCC提取法、 EPOCH提取法等。 获得语音识别参数序列后，可以将获得的语音识别参数序列以文件的形 式保存在计算机的硬盘或内存中， 另外， 也可以将每个语音控制指令对应的 数字语音信号保存在计算机的硬盘或内存中， 而且， 为了节省硬盘或内存的 空间， 可以将上述数字语音信号进行压缩后再进行保存。 最后， 配置语音识别参数序列与操作指令之间的对应关系， 而且， 为了 提高语音控制指令的识别率， 针对每个操作指令， 可以预先设置一个或多个 语音控制指令， 即对于每个操作指令， 可以输入多个语音控制指令， 分别获 得该多个语音控制指令对应的语音识别参数序列， 并分别建立该操作指令与 语音识别参数序列之间的对应关系， 即一个操作指令可以对应多个语音识别 参数序列。 图 1是根据本发明实施例的监控设备的控制方法的流程图，如图 1所示， 该方法包括以下步骤： 步骤 S102, 对接收的语音控制指令对应的数字语音信号进行特征提取， 获取该语音控制指令的语音识别参数序列； 步骤 S 104 , 在预先设置的一个或多个语音识别参数序列中查找与获取 的语音识别参数序列匹配最佳的语音识别参数序列， 例如， 可以利用动态时 间归整算法、 隐含马可夫模型进行匹配。 步骤 S106, 确定查找的语音识别参数序列所对应的操作指令， 并根据 已确定的操作指令对监控设备进行控制， 具体地， 可以向目标编码器发送操 作指令， 再由目标编码器根据操作指令对监控设备进行控制。 在客户端与监 控设备不在同一个网络中时， 可以通过中转服务器向目标编码器发送操作指 令， 具体地， 客户端向中转服务器发送操作指令， 中转服务器将操作指令转 发给目标编码器， 目标编码器艮据该操作指令对监控设备进行控制。 通过本发明实施例提供的技术方案， 利用语音指令对监控设备进行控 制，使操作人员可以通过直接输入语音控制命令的方式对监控设备进行控制， 对于操作人员来说， 该方法操作简单， 且更加形象、 直观。 图 2 是根据本发明方法实施例的监控设备的控制方法的详细处理流程 图， 如图 2所示， 该方法包括以下步骤： 步骤 S201 , 针对不同的监控设备， 需要不同的操作指令， 才艮据不同的 操作指令预先设置相应的语音控制指令， 例如分别设置云台向左旋转、 向右 旋转、 向上旋转、 向下旋转的语音控制指令。 步骤 S202, 对于每个操作指令的语音控制指令， 对语音信号进行釆样， 具体地， 语音信号的釆样频率可以为 8KHZ , 该 8KHZ适合于简单语句的识 另1 J , 对于复杂的语句， 可以选择更高的釆样频率， 样本区间为 10MS , 样本 区间包含 80个釆样点，一段语音包含多个样本，每隔 10MS计算一次短时能 量和，当釆样得到的短时能量和大于一定的阈值时则认为语音釆样已经开始， 当短时能量和小于平均能量的 1/20时认为语音釆样已经结束，获得语音釆样 信号， 并利用月永冲编码调制 （Pulse-code Modulation, 简称为 PCM ) 格式保 存每个操作指令的语音釆样信号 （即语音 PCM编码）。 步骤 S203 , 对于步骤 S202中获得的每个语音釆样信号， 提取出其对应 的特征参数， 确定语音识别参数序列， 可以利用共振峰提取法、 端点检测提 取法、 线性预测到普系数提取法、 MFCC提取法、 LSF提取法、 PLCC提取 法、 EPOCH 提取法进行特征参数的提取， 本发明实施例以线性预测编码 ( LPC ) 算法为例进行说明。 具体地， 对于每个语音釆样信号， 计算出语音釆样信号所对应的 12阶 预测倒谱系数（ LPCC ),并将获得的一系列特征参数构造成特征矢量序列（形 式如 A = { al , a2, …， ai } )， 该特征矢量序列就是语音识别参数序列， 即 语音参数模版。 步骤 S204,将得到的语音识别参数序列和步骤 S202中釆样得到的语音 PCM编码以文件的形式保存在客户端的计算机硬盘或内存中， 其中， 文件名 称应与控制指令相对应， 例如， 操作指令为监控设备向右转动， 则该文件名 称可以为 Template— Right。 另夕卜， 可以将保存的文件一次性读入硬盘或内存 中， 而不必每次识别语音命令时反复读入， 以节省识别过程的时间。 通过上述步骤 S201至步骤 S204, 实现了对与监控设备的操作指令对应 的语音识别参数序列的设置， 下面对利用语音操作指令控制监控设备进行说 明。 步骤 S205 , 输入语音控制命令， 并将该语音控制命令进行釆样， 该实 现方法与步骤 S202相同， 这里不再赘述。 步骤 S206,获得与步骤 S205中的语音控制命令对应的语音识别参数序 歹' J S = { si , s2, ··· , sk}， 该实现方法与步骤 S203相同， 这里不再赞述。 步骤 S207 ,将步骤 S206中获得的语音识别参数序列 S与预先设置的多 个语音识别参数序列进行匹配， 选择最佳匹配的语音识别参数， 并根据该最 佳匹配的语音识别参数对应的操作指令对监控设备进行控制。 例如， 本地预 先设置了四个语音识别参数序列 A = { al , a2, …， ai }、 B = { bl , b2, bj }、 C = { cl , c2, ··· , cm }、 D = { dl , d2, …， dn }， 分别对应的操作才旨 令为向左 ( Template_Left ), 向右 （ Template_Right )、 向上 （ Template_Up )、 向下 （ Template_Down ), 使用 DTW (动态时间归整） 算法， 将输入的语音 识别参数序列 S与模板库中保存的参考模板 A、 B、 C、 D依次进行匹配， 其 中匹配度最高的参考模版即为识别结果， 根据识别结果确定其所代表的控制 指令， 如语音识别参数序列 S与参考模板 A最为匹配， 则确定对监控设备执 行的操作指令为参考模板 A所对应的向左 （ Template— Left )。 步骤 S208 , 客户端与解码器建立 TCP短连接， 并向远程的编码器发送 步骤 S207中确定的监控设备控制请求 （即， 上文所述的操作指令）， 例如可 以以 XML格式发送监控设备控制请求， 该监控设备控制请求消息中可以包 括消息目的监控设备的 ID、 控制的方式、 控制的方向和控制步长量等信息， 并获取编码器的响应消息， 完成对监控设备的控制操作。 其中， 图 3示出了 实施该方法的具体实施环境， 如图 3所述， 客户端可以通过网络向目标编码 器发送监控设备控制消息， 编码器提供了实现流媒体数据的设备以响应监控 设备控制请求， 编码器与监控设备和摄像头直接相连， 在收到客户端的监控 设备控制请求后， 通过 485口向监控设备发送相应的控制指令， 完成对监控 设备的控制操作。 需要说明的是， 在本发明实施例中， 向远程的编码器发送监控设备控制 请求时， 使用的是 TCP短链的连接方式， 但并不限于此， 客户端可以根据与 编码器的协商， 釆用 TCP长链的连接方式或 UPD等其它连接方式， 而且监 控设备控制请求也可以釆用 XML之外的其它格式来实现。 另夕卜， 本发明实施例为使用语音命令控制监控设备向左、 向右、 向上、 向下旋转， 但并不限于次， 通过本发明提供的技术方案， 可以支持控制监控 设备和摄像头的其它更多的操作， 例如控制摄像头变焦、 调整亮度、 操纵监 控设备的附属设备如灯光和雨刷等。 才艮据本发明实施例， 还提供了一种计算机可读介质， 该计算机可读介质 上存储有计算机可执行的指令， 当该指令被计算机或处理器执行时， 使得计 算机或处理器执行如图 1及图 2所示的所有步骤的处理。 装置实施例 根据本发明实施例， 提供一种监控设备的控制装置。 图 4示出了才艮据本发明实施例的监控设备的控制装置的结构框架图，如 图 4所示， 该装置包括获取模块 10、 匹配模块 20和控制模块 30。 下面详细描述上述各模块。 获取模块 10, 用于对接收的语音控制指令对应的数字语音信号进行特 征提取， 获取语音控制指令的语音识别参数序列。 匹配模块 20 , 用于在预先设置的一个或多个语音识别参数序列中查找 与获取的语音识别参数序列匹配最佳的语音识别参数序列， 该模块可以连接 至获取模块 10。 控制模块 30, 用于确定查找到的语音识别参数序列所对应的操作指令， 并利用操作指令对监控设备进行控制， 该模块可以连接至匹配模块 20。 通过本发明实施例提供的监控设备的控制装置，利用语音指令对监控设 备进行控制， 使操作人员可以通过直接输入语音控制命令的方式对监控设备 进行控制， 对于操作人员来说， 该方法操作简单， 且更加形象、 直观。 图 5是根据本发明实施例的监控设备的控制装置的具体结构框架图，在 图 4所示的装置的基础上， 图 5所示的装置还包括接收模块 40、 保存模块 50和配置模块 60。 对上述各模块进行详细描述。 接收模块 40 , 用于预先接收一个或多个语音控制指令。 保存模块 50, 用于对每个语音控制指令对应的数字语音信号进行特征 提取， 获取并保存每个语音控制指令的语音识别参数序列， 该模块可以连接 至匹配模块 20和接收模块 40。 配置模块 60 , 用于配置语音识别参数序列与操作指令之间的对应关系， 该模块可以连接至保存模块 50。 其中， 保存模块 50还用于保存预先接收的数字语音信号； 或者， 保存 压缩后的每个语音控制指令对应的数字语音信号。 如上所述， 借助于本发明提供的监控设备的控制方法和 /或装置， 利用 语音指令对监控设备进行控制， 使操作人员可以通过直接输入语音控制命令 的方式对监控设备进行控制， 对于操作人员来说， 该方法操作简单， 且更加 形象、 直观。 以上所述仅为本发明的优选实施例而已， 并不用于限制本发明， 对于本 领域的技术人员来说， 本发明可以有各种更改和变^^ 凡在本发明的^^申和 原则之内， 所作的任何修改、 等同替换、 改进等， 均应包含在本发明的保护 范围之内。 The present invention relates to the field of communications, and in particular, to a method and an apparatus for controlling a monitoring device. BACKGROUND With the development of network technologies and the improvement of user security awareness, the use of video surveillance for security defense has been widely used. In video surveillance, the pan/tilt is usually used to control the camera's monitoring angle. The pan/tilt is a camera mounting platform consisting of two AC motors. The operator operates the pan/tilt to make the pan/tilt move horizontally or vertically. The camera realizes the change of the monitoring angle and performs video monitoring. At present, the implementation method of controlling the pan/tilt is: The monitoring device client sends a control command to the monitoring device by using a mouse or a keyboard to control the monitoring device. This monitoring device control method is complicated to operate, and only a professional can operate the monitoring device, and the method is not image and intuitive for the operator. SUMMARY OF THE INVENTION The present invention has been made in view of the problem of complicated operation of a monitoring device control method existing in the related art. Therefore, it is a primary object of the present invention to provide a control method and apparatus for a monitoring device to solve the above problems. According to an aspect of the invention, a method of controlling a monitoring device is provided. The control method of the monitoring device according to the present invention comprises: performing feature extraction on a digital voice signal corresponding to the received voice control instruction, acquiring a voice recognition parameter sequence of the voice control instruction; and pre-setting one or more voice recognition parameter sequences The search and the acquired speech recognition parameter sequence match the best speech recognition parameter sequence; determine the operation instruction corresponding to the searched speech recognition parameter sequence, and use the operation instruction to control the monitoring device. Before the voice control instruction is received, the method further includes: receiving one or more voice control instructions in advance; performing feature extraction on the corresponding digital voice signal for each voice control instruction, acquiring and saving each voice control Instruction speech recognition parameter sequence; configuration speech recognition The correspondence between the parameter sequence and the operation instruction. Further, the above method further includes: saving, for each voice control instruction received in advance, a corresponding digital voice signal. The operation for saving the corresponding digital voice signal for each voice control command received in advance is specifically: compressing the corresponding digital voice signal for each voice control signaling received in advance, and saving the compressed digital number voice signal. Preferably, for each operation instruction, the number of preset voice control instructions may be one or more. The operation of controlling the device by using the operation instruction is specifically: sending an operation instruction to the target encoder, and the target encoder controls the monitoring device according to the operation instruction. The operation of controlling the device by using the operation instruction is specifically: sending an operation instruction to the relay server, and the relay server forwards the operation instruction to the target encoder, and the target encoder controls the monitoring device according to the operation instruction. Preferably, the digital speech signal characteristic parameter corresponding to the voice control instruction may be extracted by using one of the following methods: a formant extraction method, an endpoint detection extraction method, a linear prediction to a general coefficient extraction method, a Mel cepstral coefficient extraction method, and a linear frequency extraction method. . Preferably, the manner of finding the best speech recognition parameter sequence matching the acquired speech recognition parameter sequence comprises at least one of the following: a dynamic time rounding algorithm, a hidden Markov model. According to another aspect of the present invention, a control device for a monitoring device is provided. The control device of the monitoring device according to the present invention includes: an obtaining module, configured to perform feature extraction on a digital voice signal corresponding to the received voice control instruction, to obtain a voice recognition parameter sequence of the voice control instruction; and a matching module, configured to be preset The one or more speech recognition parameter sequences are matched with the acquired speech recognition parameter sequence to match the best speech recognition parameter sequence; the control module is configured to determine an operation instruction corresponding to the searched speech recognition parameter sequence, and monitor the operation instruction The device is controlled. Further, the device further includes: a receiving module, configured to receive one or more voice control commands in advance; and a saving module, configured to perform feature extraction on the digital voice signal corresponding to each voice control instruction, acquire and save each voice control Instruction speech recognition parameter sequence; configuration module, for Configure the correspondence between the speech recognition parameter sequence and the operation instruction. The saving module is further configured to save the digital voice signal received in advance; or save the digital voice signal corresponding to each compressed voice control instruction. Through the above at least one technical solution of the present invention, the monitoring device is controlled by using a voice instruction, and the operator can control the monitoring device by directly inputting the voice control command, and the method is simple and more image-oriented for the operator. Intuitive. The drawings are intended to provide a further understanding of the invention, and are intended to be a part of the description of the invention. In the drawings: FIG. 1 is a flowchart of a control method of a monitoring device according to an embodiment of a method of the present invention; FIG. 2 is a detailed processing flowchart of a control method of a monitoring device according to an embodiment of a method of the present invention; FIG. 4 is a structural block diagram of a control device of a monitoring device according to an embodiment of the device of the present invention; FIG. 5 is a specific embodiment of a control device for a monitoring device according to an embodiment of the device of the present invention; Structure frame diagram. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS As described above, at present, when there is a control over a monitoring device, there is a problem that the operation is complicated. The present invention addresses a problem, and proposes a control scheme for the monitoring device, which uses a voice control command. Controlling the monitoring equipment, the scheme is simple to operate, and is more visual and intuitive than the prior art, and with the rapid development of technology, the speech recognition technology has gradually become a key technology of human-machine interface in information technology. The invention will be described in detail below with reference to the accompanying drawings. It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict. Method Embodiments According to an embodiment of the present invention, a control method of a monitoring device is provided. In the embodiment of the present invention, the correspondence between the speech recognition parameter sequence and the operation instruction needs to be pre-configured. Specifically, one or more voice control commands may be received in advance, and the analog voice signal is converted for each voice control instruction. For the digital speech signal and feature extraction of the above digital speech signal, a speech recognition parameter sequence of each speech control instruction is obtained, wherein there are various ways to extract the feature of the digital speech signal, for example, the formant extraction method can be used. , endpoint detection extraction method, linear prediction to general coefficient extraction method, Mel cepstral coefficient (MFCC) extraction method, linear frequency (LSF) extraction method, PLCC extraction method, EPOCH extraction method, and the like. After obtaining the speech recognition parameter sequence, the obtained speech recognition parameter sequence may be saved in the form of a file on the hard disk or the memory of the computer. In addition, the digital voice signal corresponding to each voice control instruction may be saved in the hard disk or the memory of the computer. In addition, in order to save space on the hard disk or the memory, the digital voice signal can be compressed and then saved. Finally, the correspondence between the speech recognition parameter sequence and the operation instruction is configured, and, in order to improve the recognition rate of the voice control instruction, one or more voice control instructions may be preset for each operation instruction, that is, for each operation instruction a plurality of voice control commands may be input, respectively, and the voice recognition parameter sequences corresponding to the plurality of voice control commands are respectively obtained, and a correspondence relationship between the operation command and the voice recognition parameter sequence is respectively established, that is, one operation command may correspond to multiple voices. Identify the sequence of parameters. 1 is a flowchart of a method for controlling a monitoring device according to an embodiment of the present invention. As shown in FIG. 1, the method includes the following steps: Step S102: Perform feature extraction on a digital voice signal corresponding to a received voice control command, and obtain the a voice recognition parameter sequence of the voice control command; Step S104, searching for a sequence of the voice recognition parameter that matches the obtained voice recognition parameter sequence in the preset one or more voice recognition parameter sequences, for example, using dynamic time return The whole algorithm, the implicit Markov model, is matched. Step S106, determining an operation instruction corresponding to the searched speech recognition parameter sequence, and controlling the monitoring device according to the determined operation instruction. Specifically, the operation instruction may be sent to the target encoder, and then the target encoder is configured according to the operation instruction. The monitoring device is controlled. In the client and supervisor When the control device is not in the same network, the operation command may be sent to the target encoder through the relay server. Specifically, the client sends an operation instruction to the relay server, and the relay server forwards the operation instruction to the target encoder, and the target encoder uses the The operating instructions control the monitoring device. Through the technical solution provided by the embodiment of the present invention, the monitoring device is controlled by using a voice instruction, so that the operator can control the monitoring device by directly inputting the voice control command, and the method is simple and more convenient for the operator. Image, intuitive. 2 is a detailed processing flowchart of a method for controlling a monitoring device according to an embodiment of the method of the present invention. As shown in FIG. 2, the method includes the following steps: Step S201: Different operating instructions are required for different monitoring devices. Corresponding voice control commands are preset according to different operation commands, for example, voice control commands for setting the pan/tilt to rotate leftward, rightward, upward, and downward. Step S202, the operation instruction for each voice control commands, Bian speech signal samples, particular, preclude the sampling frequency of the speech signal may be 8KHZ, the 8KHZ suitable for simple statement identifying another 1 J, for complex statements, You can choose a higher sampling frequency, the sample interval is 10MS, the sample interval contains 80 sample points, a segment of speech contains multiple samples, and the short-term energy sum is calculated every 10MS, and the short-term energy obtained by the sample is greater than At a certain threshold, it is considered that the voice sample has begun. When the short-term energy is less than 1/20 of the average energy, the voice sample is considered to have ended, the voice sample signal is obtained, and the pulse-code modulation is utilized. , abbreviated as PCM) format to store the voice sample signal (ie, voice PCM code) for each operation command. Step S203, extracting corresponding feature parameters for each of the voice sample signals obtained in step S202, and determining a sequence of voice recognition parameters, which may use a formant extraction method, an endpoint detection extraction method, a linear prediction to a general coefficient extraction method, The feature parameters are extracted by the MFCC extraction method, the LSF extraction method, the PLCC extraction method, and the EPOCH extraction method. The linear prediction coding (LPC) algorithm is taken as an example for description. Specifically, for each voice sample signal, a 12th-order predictive cepstral coefficient (LPCC) corresponding to the voice sample signal is calculated, and the obtained series of feature parameters are constructed into a feature vector sequence (form as A = { al , a2, ..., ai } ), the feature vector sequence is a speech recognition parameter sequence, that is, a speech parameter template. Step S204, the obtained voice recognition parameter sequence and the voice PCM code obtained in step S202 are saved in the form of a file on a computer hard disk or a memory of the client, where the file name The name should correspond to the control instruction. For example, if the operation instruction is to rotate the monitoring device to the right, the file name can be Template_Right. In addition, the saved file can be read into the hard disk or the memory at one time without having to repeatedly read in each time the voice command is recognized, thereby saving the time of the recognition process. Through the above steps S201 to S204, the setting of the speech recognition parameter sequence corresponding to the operation instruction of the monitoring device is realized, and the following uses the voice operation command to control the monitoring device. Step S205, the voice control command is input, and the voice control command is sampled. The implementation method is the same as step S202, and details are not described herein again. Step S206, the speech recognition parameter sequence 歹' JS = { si , s2, ···, sk} corresponding to the voice control command in step S205 is obtained, and the implementation method is the same as step S203, and is not mentioned here. Step S207, matching the speech recognition parameter sequence S obtained in step S206 with a plurality of preset speech recognition parameter sequences, selecting the best matching speech recognition parameter, and according to the operation instruction corresponding to the best matching speech recognition parameter Control the monitoring equipment. For example, four speech recognition parameter sequences A = { al , a2, ..., ai }, B = { bl , b2, bj }, C = { cl , c2, ··· , cm }, D = are preset in the local area. { dl , d2 , ..., dn }, respectively, the corresponding operations are left (Template_Left), right (Template_Right), up (Template_Up), down (Template_Down), using DTW (Dynamic Time Rounding) algorithm The input speech recognition parameter sequence S is sequentially matched with the reference templates A, B, C, and D stored in the template library, wherein the reference template with the highest matching degree is the recognition result, and the control instruction represented by the recognition result is determined according to the recognition result. If the speech recognition parameter sequence S matches the reference template A most, it is determined that the operation instruction executed on the monitoring device is the leftward (Template_Left) corresponding to the reference template A. Step S208, the client establishes a TCP short connection with the decoder, and sends the monitoring device control request (ie, the operation instruction described above) determined in step S207 to the remote encoder, for example, the monitoring device control may be sent in an XML format. The request, the monitoring device control request message may include information such as the ID of the monitoring device of the message destination, the manner of control, the direction of the control, and the amount of control step, and obtains the response message of the encoder to complete the control operation of the monitoring device. FIG. 3 shows a specific implementation environment for implementing the method. As shown in FIG. 3, the client may send a monitoring device control message to the target encoder through the network, and the encoder provides a device for implementing streaming media data in response to the monitoring device. The control request is directly connected to the monitoring device and the camera. After receiving the control device control request from the client, the 485 port sends a corresponding control command to the monitoring device to complete the control operation of the monitoring device. It should be noted that, in the embodiment of the present invention, when the monitoring device control request is sent to the remote encoder, the TCP short-chain connection mode is used, but the method is not limited thereto, and the client may negotiate according to the encoder. TCP TCP long-chain connection or other connection methods such as UPD, and monitoring device control requests can also be implemented in formats other than XML. In addition, the embodiment of the present invention uses a voice command to control the monitoring device to rotate left, right, up, and down, but is not limited to the second, and the technical solution provided by the present invention can support other devices for controlling the monitoring device and the camera. More operations, such as controlling camera zoom, adjusting brightness, and manipulating the monitoring equipment's accessories such as lights and wipers. According to an embodiment of the present invention, there is also provided a computer readable medium having stored thereon computer executable instructions for causing a computer or processor to perform, for example, when executed by a computer or processor The processing of all the steps shown in Figures 1 and 2. Apparatus Embodiment According to an embodiment of the present invention, a control apparatus for a monitoring apparatus is provided. 4 is a structural block diagram of a control device of a monitoring device according to an embodiment of the present invention. As shown in FIG. 4, the device includes an acquisition module 10, a matching module 20, and a control module 30. The above modules are described in detail below. The obtaining module 10 is configured to perform feature extraction on the digital voice signal corresponding to the received voice control instruction, and obtain a voice recognition parameter sequence of the voice control instruction. The matching module 20 is configured to search for a sequence of the best speech recognition parameter that matches the acquired sequence of the speech recognition parameter in the preset one or more speech recognition parameter sequences, and the module may be connected to the acquisition module 10. The control module 30 is configured to determine an operation instruction corresponding to the searched speech recognition parameter sequence, and control the monitoring device by using an operation instruction, and the module may be connected to the matching module 20. The control device of the monitoring device provided by the embodiment of the present invention controls the monitoring device by using a voice instruction, so that the operator can control the monitoring device by directly inputting the voice control command, and the operation is simple for the operator. , and more visual and intuitive. 5 is a detailed structural diagram of a control device of a monitoring device according to an embodiment of the present invention, Based on the apparatus shown in FIG. 4, the apparatus shown in FIG. 5 further includes a receiving module 40, a saving module 50, and a configuration module 60. The above modules are described in detail. The receiving module 40 is configured to receive one or more voice control commands in advance. The saving module 50 is configured to perform feature extraction on the digital voice signal corresponding to each voice control instruction, acquire and save a voice recognition parameter sequence of each voice control instruction, and the module may be connected to the matching module 20 and the receiving module 40. The configuration module 60 is configured to configure a correspondence between the speech recognition parameter sequence and the operation instruction, and the module may be connected to the saving module 50. The saving module 50 is further configured to save the digital voice signal received in advance; or save the digital voice signal corresponding to each compressed voice control instruction. As described above, with the control method and/or device of the monitoring device provided by the present invention, the monitoring device is controlled by using a voice instruction, so that the operator can control the monitoring device by directly inputting the voice control command, for the operator In other words, the method is simple, and more visual and intuitive. The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention. For those skilled in the art, the present invention can be variously modified and modified. Any modifications, equivalent substitutions, improvements, etc. made therein are intended to be included within the scope of the present invention.

Claims

权 利 要 求 书 Claim
1. 一种监控设备的控制方法， 其特征在于， 包括： A control method for a monitoring device, comprising:
对接收的语音控制指令对应的数字语音信号进行特征提取，获取所 述语音控制指令的语音识别参数序列；  Performing feature extraction on the digital voice signal corresponding to the received voice control instruction, and acquiring a voice recognition parameter sequence of the voice control instruction;
在预先设置的一个或多个语音识别参数序列中查找与所述获取的 语音识别参数序列匹配最佳的语音识别参数序列；  Locating a sequence of speech recognition parameters that match the acquired sequence of speech recognition parameters in a sequence of one or more speech recognition parameters set in advance;
确定所述查找的语音识别参数序列所对应的操作指令，并利用所述 操作指令对监控设备进行控制。  Determining an operation instruction corresponding to the searched speech recognition parameter sequence, and using the operation instruction to control the monitoring device.
2. 根据权利要求 1所述的方法， 其特征在于， 在接收到所述语音控制指令 之前， 所述方法还包括： The method according to claim 1, wherein before the receiving the voice control instruction, the method further includes:
预先接收一个或多个语音控制指令；  Receiving one or more voice control commands in advance;
对于每个语音控制指令， 将其对应的数字语音信号进行特征提取， 获取并保存每个语音控制指令的语音识别参数序列； 配置语音识别参数序列与所述操作指令之间的对应关系。  For each voice control instruction, feature extracting the corresponding digital voice signal, acquiring and storing a voice recognition parameter sequence of each voice control instruction; and configuring a correspondence relationship between the voice recognition parameter sequence and the operation instruction.
3. 根据权利要求 2所述的方法， 其特征在于， 所述方法还包括： The method according to claim 2, wherein the method further comprises:
对于预先接收的所述每个语音控制指令，保存其对应的数字语音信 号。  For each of the voice control commands received in advance, the corresponding digital voice signal is saved.
4. 根据权利要求 3所述的方法， 其特征在于， 对于预先接收的所述每个语 音控制指令， 保存其对应的数字语音信号的操作具体为： The method according to claim 3, wherein the operation of saving the corresponding digital voice signal for each of the voice control instructions received in advance is:
对于预先接收的所述每个语音控制信令，压缩其对应的数字语音信 号， 并保存压缩后的所述数字语音信号。  For each of the voice control signaling received in advance, the corresponding digital voice signal is compressed, and the compressed digital voice signal is saved.
5. 根据权利要求 2所述的方法， 其特征在于， 对于每个操作指令， 预先设 置的语音控制指令的数量为一个或多个。 The method according to claim 2, wherein the number of preset voice control instructions is one or more for each operation instruction.
6. 根据权利要求 1所述的方法， 其特征在于， 利用所述操作指令对设备进 行控制的操作具体为： The method according to claim 1, wherein the operation of controlling the device by using the operation instruction is specifically:
向目标编码器发送所述操作指令，所述目标编码器根据所述操作指 令对所述监控设备进行控制。 The operation instruction is sent to a target encoder, and the target encoder controls the monitoring device according to the operation instruction.
7. 根据权利要求 1所述的方法， 其特征在于， 利用所述操作指令对设备进 行控制的操作具体为： The method according to claim 1, wherein the operation of controlling the device by using the operation instruction is specifically:
向中转服务器发送所述操作指令，所述中转服务器将所述操作指令 转发给目标编码器， 所述目标编码器艮据所述操作指令对所述监控设备 进行控制。  Sending the operation instruction to the relay server, the relay server forwards the operation instruction to the target encoder, and the target encoder controls the monitoring device according to the operation instruction.
8. 才艮据权利要求 1至 7中任一项所述的方法， 其特征在于， 利用以下方式 之一提取所述语音控制指令对应的数字语音信号特征参数： 共振峰提取 法、端点检测提取法、线性预测倒 i普系数提取法、 Mel倒 i普系数即 MFCC 提取法、 线性频率即 LSF提取法。 The method according to any one of claims 1 to 7, characterized in that the digital speech signal characteristic parameter corresponding to the voice control instruction is extracted by one of the following methods: formant extraction method, endpoint detection extraction Method, linear prediction, inverse factor extraction method, Mel inverted coefficient, MFCC extraction method, linear frequency, LSF extraction method.
9. 才艮据权利要求 1至 7中任一项所述的方法， 其特征在于， 查找与所述获 取的语音识别参数序列匹配最佳的语音识别参数序列的方式包括以下至 少之一： 动态时间归整算法、 隐含马可夫模型。 9. The method according to any one of claims 1 to 7, characterized in that the manner of finding the best speech recognition parameter sequence matching the acquired speech recognition parameter sequence comprises at least one of the following: Time rounding algorithm, implicit Markov model.
10. 一种监控设备的控制装置， 其特征在于， 包括： A control device for a monitoring device, comprising:
获取模块，用于对接收的语音控制指令对应的数字语音信号进行特 征提取， 获取所述语音控制指令的语音识别参数序列；  And an acquiring module, configured to perform feature extraction on the digital voice signal corresponding to the received voice control instruction, and acquire a sequence of voice recognition parameters of the voice control instruction;
匹配模块，用于在预先设置的一个或多个语音识别参数序列中查找 与所述获取的语音识别参数序列匹配最佳的语音识别参数序列；  a matching module, configured to search, in a preset one or more speech recognition parameter sequences, a sequence of speech recognition parameters that matches an optimal sequence of the acquired speech recognition parameter;
控制模块，用于确定所述查找的语音识别参数序列所对应的操作指 令， 并利用所述操作指令对监控设备进行控制。  And a control module, configured to determine an operation instruction corresponding to the searched voice recognition parameter sequence, and use the operation instruction to control the monitoring device.
11. 才艮据权利要求 10所述的装置， 其特征在于， 所述装置还包括： 11. The device according to claim 10, wherein the device further comprises:
接收模块， 用于预先接收一个或多个语音控制指令； 保存模块，用于对每个语音控制指令对应的数字语音信号进行特征 提取， 获取并保存每个语音控制指令的语音识别参数序列；  a receiving module, configured to receive one or more voice control commands in advance; a saving module, configured to perform feature extraction on a digital voice signal corresponding to each voice control instruction, and acquire and save a voice recognition parameter sequence of each voice control instruction;
配置模块，用于配置语音识别参数序列与所述操作指令之间的对应 关系。  And a configuration module, configured to configure a correspondence between the sequence of the speech recognition parameter and the operation instruction.
12. 根据权利要求 10或 11所述的装置， 其特征在于， 所述保存模块还用于 保存预先接收的数字语音信号； 或者， 保存压缩后的所述每个语音控制 指令对应的数字语音信号。 The device according to claim 10 or 11, wherein the saving module is further configured to save a digital voice signal received in advance; or save the compressed digital voice signal corresponding to each voice control command .
PCT/CN2009/072503 2008-08-22 2009-06-29 Control method and device for monitoring equipment WO2010020138A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200810214403.0 2008-08-22
CN 200810214403 CN101345668A (en) 2008-08-22 2008-08-22 Control method and apparatus for monitoring equipment

Publications (1)

Publication Number Publication Date
WO2010020138A1 true WO2010020138A1 (en) 2010-02-25

Family

ID=40247563

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/072503 WO2010020138A1 (en) 2008-08-22 2009-06-29 Control method and device for monitoring equipment

Country Status (2)

Country Link
CN (1) CN101345668A (en)
WO (1) WO2010020138A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102148032A (en) * 2010-12-03 2011-08-10 北京声迅电子有限公司 Abnormal sound detection method and system for ATM (Automatic Teller Machine)
CN117893244A (en) * 2024-03-15 2024-04-16 中国海洋大学 Comprehensive management and control system for seaweed hydrothermal carbonization application based on machine learning

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101345668A (en) * 2008-08-22 2009-01-14 中兴通讯股份有限公司 Control method and apparatus for monitoring equipment
CN101478666A (en) * 2009-01-19 2009-07-08 北京中星微电子有限公司 Tripod head control task operating method and apparatus thereof for terminal video taking device
CN102375535B (en) * 2010-08-25 2015-11-25 北京中星微电子有限公司 Man-machine interaction method in a kind of video monitoring and device, video monitoring system
CN102023644A (en) * 2010-11-10 2011-04-20 新太科技股份有限公司 Method for controlling cradle head based on voice recognition technology
CN103297590B (en) * 2012-02-24 2016-12-14 中兴通讯股份有限公司 A kind of method and system realizing equipment unblock based on audio frequency
CN104766608A (en) * 2014-01-07 2015-07-08 深圳市中兴微电子技术有限公司 Voice control method and voice control device
CN104092936B (en) * 2014-06-12 2017-01-04 小米科技有限责任公司 Atomatic focusing method and device
CN104064021B (en) * 2014-06-20 2018-07-27 Tcl集团股份有限公司 A kind of learning method of remote controler, device and amusement audio-visual equipment
CN104103272B (en) * 2014-07-15 2017-10-10 无锡中感微电子股份有限公司 Audio recognition method, device and bluetooth earphone
CN104135619A (en) * 2014-08-12 2014-11-05 广东欧珀移动通信有限公司 Method and device of controlling camera
CN105141919A (en) * 2015-09-01 2015-12-09 武汉同迅智能科技有限公司 Monitoring terminal device remotely controlled by voice
CN105681753A (en) * 2016-01-18 2016-06-15 阿亦睿机器人科技（上海）有限公司 Intelligent monitoring device
CN106534807A (en) * 2016-12-19 2017-03-22 重庆旭穆科技有限公司 Control method of monitoring device
CN107146616B (en) 2017-06-13 2020-05-08 Oppo广东移动通信有限公司 Equipment control method and related product
CN109963073A (en) * 2017-12-26 2019-07-02 浙江宇视科技有限公司 Video camera control method, device, system and PTZ camera
CN108831467A (en) * 2018-07-18 2018-11-16 深圳市普威技术有限公司 A kind of IPC equipment based on voice control
CN110827811A (en) * 2018-08-13 2020-02-21 格力电器（武汉）有限公司 Voice control method and device for household electrical appliance
CN110556106A (en) * 2019-07-30 2019-12-10 视联动力信息技术股份有限公司 monitoring service processing method and system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1713717A (en) * 2004-06-25 2005-12-28 北京中星微电子有限公司 Digital sound control orienting method for camera site of camera
CN101038742A (en) * 2006-03-16 2007-09-19 普立尔科技股份有限公司 Apparatus and method for assistant voice remote control using image feature
CN101345668A (en) * 2008-08-22 2009-01-14 中兴通讯股份有限公司 Control method and apparatus for monitoring equipment

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1713717A (en) * 2004-06-25 2005-12-28 北京中星微电子有限公司 Digital sound control orienting method for camera site of camera
CN101038742A (en) * 2006-03-16 2007-09-19 普立尔科技股份有限公司 Apparatus and method for assistant voice remote control using image feature
CN101345668A (en) * 2008-08-22 2009-01-14 中兴通讯股份有限公司 Control method and apparatus for monitoring equipment

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102148032A (en) * 2010-12-03 2011-08-10 北京声迅电子有限公司 Abnormal sound detection method and system for ATM (Automatic Teller Machine)
CN117893244A (en) * 2024-03-15 2024-04-16 中国海洋大学 Comprehensive management and control system for seaweed hydrothermal carbonization application based on machine learning
CN117893244B (en) * 2024-03-15 2024-06-04 中国海洋大学 Comprehensive management and control system for seaweed hydrothermal carbonization application based on machine learning

Also Published As

Publication number Publication date
CN101345668A (en) 2009-01-14

Similar Documents

Publication Publication Date Title
WO2010020138A1 (en) Control method and device for monitoring equipment
JP6613347B2 (en) Method and apparatus for pushing information
US20210166699A1 (en) Methods and apparatus for hybrid speech recognition processing
EP3195310B1 (en) Keyword detection using speaker-independent keyword models for user-designated keywords
JP2021516790A (en) System and method of selective wake word detection using neural network model
CN111279296B (en) Creating modular dialogs using implicit routing
US20160103825A1 (en) Mobile speech-to-speech interpretation system
EP3197139B1 (en) Information processing system, information processing device, and information processing method
EP2960901A1 (en) User-adapted speech recognition
US20020198714A1 (en) Statistical spoken dialog system
US20160125880A1 (en) Method and system for identifying location associated with voice command to control home appliance
EP3916538B1 (en) Creating a cinematic storytelling experience using network-addressable devices
WO2019096056A1 (en) Speech recognition method, device and system
CN102568478A (en) Video play control method and system based on voice recognition
WO2020233363A1 (en) Speech recognition method and device, electronic apparatus, and storage medium
WO2015103836A1 (en) Voice control method and device
JPWO2018100743A1 (en) Control device and equipment control system
KR20230107860A (en) Voice personalization and federation training using real noise
CN109600424A (en) A kind of integrated mainframe micro, network insertion and audio collection classroom wisdom terminal
WO2019101099A1 (en) Video program identification method and device, terminal, system, and storage medium
CN109688271A (en) The method, apparatus and terminal device of contact information input
JP7192348B2 (en) Control device, control system and program
CN116863910A (en) Speech data synthesis method and device, electronic equipment and storage medium
CN112700770A (en) Voice control method, sound box device, computing device and storage medium
CN112218137B (en) Multimedia data acquisition method, device, equipment and medium

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: 09807837

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09807837

Country of ref document: EP

Kind code of ref document: A1