WO2016112665A1 - 一种语音数据处理方法和装置 - Google Patents
一种语音数据处理方法和装置 Download PDFInfo
- Publication number
- WO2016112665A1 WO2016112665A1 PCT/CN2015/083943 CN2015083943W WO2016112665A1 WO 2016112665 A1 WO2016112665 A1 WO 2016112665A1 CN 2015083943 W CN2015083943 W CN 2015083943W WO 2016112665 A1 WO2016112665 A1 WO 2016112665A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- customer
- name
- data
- user
- recording
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
- H04L65/40—Support for services or applications
Definitions
- the present invention relates to the field of data processing, and more particularly to a voice data processing method and apparatus.
- the present invention provides a voice data processing method and apparatus, which converts the recording data of the promotion data and the customer into text data, and converts the processing of the recorded data from complex listening to convenient viewing.
- a voice data processing method is provided, and the method may include:
- Converting the recording file into a text file extracting one or more keywords related to the customer as customer data of the customer from the converted text file, the keyword including the customer's unit name and attention information.
- the method may further include: comparing the extracted unit name of the customer with a registered name in an administrative authority database to obtain a registered name of the unit where the customer is located as the customer
- the first supplementary part of the data wherein the administrative agency database includes one or more of a database of industrial and commercial organizations and a database of civil affairs organs.
- the method may further include: comparing the extracted unit name of the customer with a business name captured from a webpage that obtains the customer communication manner to obtain the customer location
- the page name of the unit serves as the second supplement to the customer profile.
- the method may further comprise: displaying the customer profile to the first user in text form via a browser plugin and uploading the customer profile to a server.
- the method may further include: responding to the customer profile request of the second user, and transmitting the client identifier to the second user according to the client identifier in the client profile resolution request Customer information such that the second user communicates with the customer again based on the customer profile and updates the customer profile based on the re-communication.
- a voice data processing apparatus may include:
- a data exchange module configured to respond to a recording resolution request initiated by the first user, and obtain a recording file of the client corresponding to the customer identifier according to the customer identifier in the recording resolution request,
- a voice recognition module configured to convert the sound recording file into a text file, and extract one or more keywords related to the customer as customer data of the customer from the converted text file, the keyword including the Customer's organization name and attention information.
- the voice recognition module may be further configured to compare the extracted unit name of the client with a registered name in an administrative database to obtain a registered name of the client's organization as a location
- the first supplementary part of the customer data wherein the administrative agency database includes one or more of a database of industrial and commercial organizations and a database of civil affairs organs.
- the voice recognition module may be further configured to compare the extracted unit name of the client with a company name captured from a webpage that obtains the customer communication manner.
- the webpage name of the unit where the customer is located is obtained as the second supplementary part of the customer data.
- the data exchange module may be further configured to display the customer profile to the first user in a text form through a browser plug-in, and upload the customer profile to a server.
- the server may be further configured to respond to the second user's customer profile resolution request, and send the client identifier to the second user according to the client identifier in the client profile resolution request.
- Customer information such that the second user communicates with the customer again based on the customer profile and updates the customer profile based on the re-communication.
- embodiments of the present invention provide a system, which may include: one or more processors; a memory; one or more programs, the one or more programs being stored in the memory; the one or more The operations of the voice data processing method according to the above are performed when the processor executes the one or more programs.
- Embodiments of the present invention also provide a non-volatile computer storage medium storing one or more programs when the one or more programs are executed by one or more devices And causing the one or more devices to perform operations according to the above-described voice data processing method.
- the voice data processing method and apparatus improve the information of the voice data in the communication by converting the recorded data communicated with the client into text data, and extracting the key information of the client as the customer data based on the converted text data.
- the utilization rate improves the usability of the recorded data, thereby facilitating the improvement and management of the customer data.
- the unit name based on the converted text data is compared with the official database to obtain the accurate registration name of the client unit and/or
- the unit name based on the converted text data extraction is compared with the name of the enterprise captured in the webpage for obtaining the customer communication method, and the webpage name of the client unit is obtained, thereby further improving the utilization rate of the unit information in the recorded data and improving the customer data.
- Accuracy and coverage again, displaying the customer data in text form to the user through the browser plug-in, improving the convenience of the promoters to obtain customer information, enabling the promoters to obtain customer information in real time or in time, to understand customer needs, and also to help Improve communication efficiency;
- By uploading the customer data to the server it is convenient for the promotion personnel to obtain the customer information obtained by other promoters in the first communication, expanding the use of the recorded data, improving the information sharing degree of the recorded data, and being able to pass through multiple promoters. Communication and timely update of customer information.
- FIG. 1 is a schematic flow chart showing a voice data processing method according to an embodiment of the present invention
- FIG. 2 is a schematic diagram showing the structure of a voice data processing apparatus according to an embodiment of the present invention.
- FIG. 1 a schematic flowchart of a voice data processing method according to an embodiment of the present invention is illustrated.
- the method may specifically include the following steps:
- Step S101 Respond to the recording resolution request initiated by the first user, and obtain the recording file of the customer corresponding to the customer identifier according to the customer identifier in the recording resolution request.
- Step S102 Convert the recording file into a text file, and extract one or more keywords related to the customer as the customer data of the customer from the converted text file, wherein the keyword includes the unit name and the attention information of the customer.
- the voice data processing method may include the step S101, in response to the recording resolution request initiated by the first user, acquiring the recording file of the client corresponding to the customer identifier according to the customer identifier in the recording resolution request.
- the first user may be an extension person, and the promotion personnel obtains the communication mode of the customer through a webpage or other means (for example, a phone number, a mobile phone number, etc.), and performs voice communication with the customer according to the obtained communication mode, and performs voice communication. recording.
- the promotion personnel may initiate a real-time recording resolution request with the current call client through the client (for example, a client installed on the promotion computing device) plug-in, and the recording resolution request may include Customer ID (for example, it can be the customer's phone number or mobile number, etc.).
- the promotion personnel can initiate a previous recording resolution request with a certain client through the client plug-in. For example, the promotion personnel A communicates with the customer X and performs a communication process. Recording, since the customer's intention of the demand at this time is not clear, real-time recording analysis is not requested.
- the promotion staff B After a period of time, the promotion staff B also obtains the communication method of the customer X (for example, a phone number or a mobile phone number, etc.), then the promotion person B can check the customer after the inquiry to the recording file server stores the recording file of the customer X.
- the communication method initiates a non-real-time recording resolution request as a customer identification.
- Step S101 is executed to respond to the real-time/non-real-time recording resolution request initiated by the promoter, and obtain the client identifier (for example, a phone number or a mobile phone number) in the recording resolution request from the server storing the recording file (for example, by downloading).
- the recording file of the customer corresponding to the customer identification.
- step S102 is executed to convert the acquired recording file into a text file.
- deep neural network technology can be used to learn the acoustic characteristics of the recorded data during the conversion process to optimize the recorded data.
- one or more keywords related to the customer are extracted from the converted text file as customer data of the customer, wherein one or more keywords may include the customer's unit name and attention information, and the attention information may be based on
- the related factors such as the product or the customer may be different, for example, but not limited to: promotion of the product name, product model, product price, product origin, transportation cost, and the like.
- the customer data may include the customer's unit name and attention information extracted from the converted text data.
- the customer data may also include recording information such as the date and time of the voice communication between the promoter and the customer.
- the recording data is recorded by voice recognition due to various interference factors (for example, call background noise, voice style of the two parties, etc.), wherein the availability of the customer information is low and the means for obtaining the information is time consuming. Converting to text data, and extracting customer's key information as customer data based on the converted text data, improving information utilization and convenience of voice data in communication, improving the usability of recorded data, thereby facilitating customer data Perfect and management.
- various interference factors for example, call background noise, voice style of the two parties, etc.
- the voice data processing method may further include: comparing the extracted unit name of the client with the registered name in the administrative database to obtain the registered name of the client's unit as the customer data.
- the administrative agency database includes one or more of the database of industrial and commercial organizations and the database of civil affairs organs.
- the customer unit name extracted from the converted text data is compared with the registered name in the administrative organization database.
- the registered company name in the database of the industrial and commercial organization may be compared.
- the invention can further improve the utilization of unit information in the recorded data and improve the accuracy of the customer data by comparing the above-mentioned big data with the underlying.
- the voice data processing method may further include: extracting the extracted The name of the customer's unit is compared with the name of the company captured from the webpage that obtained the communication method of the customer to obtain the name of the website of the customer's unit as the second supplementary part of the customer's data.
- the unit name extracted from the converted text data can be compared with the name of the enterprise captured in the webpage of the communication method (for example, a phone number or a mobile phone number) that is searched for the customer, thereby obtaining the customer.
- the name of the web page for example, may be an abbreviation of the enterprise or a name that is easily attracting social attention, etc., as another supplemental part of the customer's customer profile, and the supplemental portion is stored in the customer's database together with other parts of the customer profile. in.
- the voice data processing method may further include: displaying the customer profile to the first user in a text form through a browser plugin, and uploading the client profile to the server.
- the customer data in the form of text extracted based on the converted text data is displayed to the promoters in a floating layer through the browser plug-in, which improves the convenience of using the recorded data and also helps to improve the success rate of the promotion.
- the acquired customer data can also be uploaded to a Customer Relationship Management (CRM) server for easy customer data management and maintenance.
- CRM Customer Relationship Management
- the voice data processing method may further include: responding to the second user's customer profile resolution request, and transmitting the customer profile corresponding to the customer identifier to the second user according to the customer identifier in the customer profile resolution request, so that The second user communicates with the customer again based on the customer profile and updates the customer profile based on the communication again.
- the second user may be another user different from the first user (eg, another promotion person).
- the communication method can be used as a customer identifier to query whether the customer information of the customer is stored on the CRM server. When you arrive, you can directly obtain the customer's information from the CRM server. Knowing the customer's needs before communicating again can help improve communication efficiency. It is also possible to update the customer's customer profile based on the re-communicated recording data.
- the voice data processing apparatus 200 may specifically include:
- the data exchange module 201 is configured to: in response to the recording resolution request initiated by the first user, acquire a recording file of the customer corresponding to the customer identifier according to the customer identifier in the recording resolution request,
- the voice recognition module 202 is configured to convert the sound recording file into a text file, and extract one or more keywords related to the customer as customer data of the customer from the converted text file, where the keyword includes The customer's unit name and attention information.
- the modules of the voice data processing device 200 can be respectively disposed on different devices that are physically separated but communicable.
- the data exchange module 201 can be disposed in the data exchange server, and the voice recognition module 202 can be configured.
- the voice server cluster On the voice server cluster, the data exchange server and the voice server cluster can be located in physically separate locations, and each module can be called with other modules on the server device.
- the data exchange module 201 of the present invention obtains a recording file of the customer corresponding to the customer identifier according to the customer identifier in the recording resolution request in response to the recording resolution request initiated by the first user.
- the first user may be an extension person, and the promotion personnel obtains the communication mode of the customer through a webpage or other means (for example, a phone number, a mobile phone number, etc.), and performs voice communication with the customer according to the obtained communication mode, and performs voice communication. recording.
- the data exchange module 201 may initiate a real-time recording resolution request with the current call client through the client (eg, a client installed on the promotion computing device) plug-in, and the recording resolution request may include Customer ID (for example, it can be the customer's phone number or mobile number, etc.).
- the data exchange module 201 can respond to the promotion personnel to initiate a prior non-real-time recording and parsing request with a certain client through the client plug-in.
- the promotion personnel A communicates with the customer X and performs a communication process. Recording was performed, and since the customer's intention of the demand at this time was not clear, real-time recording analysis was not requested.
- the promotion staff B After a period of time, the promotion staff B also obtains the communication method of the customer X (for example, a phone number or a mobile phone number, etc.), then the promotion person B can check the customer after the inquiry to the recording file server stores the recording file of the customer X.
- the communication method initiates a non-real-time recording resolution request as a customer identification.
- the data exchange module 201 can respond to the real-time/non-real-time recording resolution request initiated by the promoter and obtain the customer identifier (for example, a phone number or a mobile phone number) in the recording resolution request from the server storing the recording file (for example, by downloading The recording file of the customer corresponding to the customer identification, and then the acquired recording file is sent to the speech recognition module 202.
- the speech recognition module 202 converts the acquired recording file into a text file.
- the deep neural network technology can be used to learn the acoustic characteristics of the recorded data during the conversion process to optimize the recorded data and extract from the converted text file.
- One or more keywords related to the customer as the customer profile of the customer may include the unit name and the attention information of the customer, and the attention information may be different according to related factors such as the product or the customer, for example, may include but not limited to: promotion product name, product model, product price, product origin Transportation costs, etc.
- the customer data may include the customer's unit name and attention information extracted from the converted text data, and the customer data may include, in addition to the above content, record information such as the date and time of the voice communication between the promoter and the customer.
- the recording data is recorded by voice recognition due to various interference factors (for example, call background noise, voice style of the two parties, etc.), wherein the availability of the customer information is low and the means for obtaining the information is time consuming. Converting to text data, and extracting customer's key information as customer data based on the converted text data, improving information utilization and convenience of voice data in communication, improving the usability of recorded data, thereby facilitating customer data Perfect and management.
- various interference factors for example, call background noise, voice style of the two parties, etc.
- the voice recognition module 202 is further configured to compare the extracted unit name of the customer with the registered name in the administrative database to obtain the registered name of the unit where the customer is located as the first of the customer data.
- the administrative agency database includes one or more of the database of industrial and commercial organizations and the database of civil affairs organs.
- the customer unit name extracted from the converted text data is compared with the registered name in the administrative organization database.
- the registered company name in the database of the industrial and commercial organization may be compared.
- the invention can further improve the utilization of unit information in the recorded data and improve the accuracy of the customer data by comparing the above-mentioned big data with the underlying.
- the voice recognition module 202 may be further configured to compare the extracted unit name of the customer with the name of the enterprise captured from the webpage that obtains the customer communication manner, to obtain the webpage name of the unit where the customer is located.
- the unit name extracted from the converted text data can be compared with the name of the enterprise captured in the webpage of the communication method (for example, a phone number or a mobile phone number) that is searched for the customer, thereby obtaining the customer.
- the name of the web page for example, may be an abbreviation of the enterprise or a name that is easily attracting social attention, etc., as another supplemental part of the customer's customer profile, and the supplemental portion is stored in the customer's database together with other parts of the customer profile. in.
- the data exchange module 201 is further configured to display the customer profile to the first user in text form through a browser plug-in, and upload the customer profile to the server.
- the customer data in the form of text extracted based on the converted text data is displayed to the promoters in a floating layer through the browser plug-in, which improves the convenience of using the recorded data and also helps to improve the success rate of the promotion.
- the acquired customer data can also be uploaded to the customer relationship management CRM server for easy customer data management and maintenance.
- the CRM server that receives the customer profile uploaded by the data exchange module 201 is further configured to: respond to the second user's customer profile resolution request, and send the second identity to the second user according to the customer identifier in the client profile resolution request.
- the customer identifies the corresponding customer profile so that the second user can communicate with the customer again based on the customer profile and update the customer profile based on the communication again.
- the second user may be another user different from the first user (eg, another promotion person).
- the promoter searches for a communication method (for example, a phone number, a mobile phone number, etc.) from a web page
- the communication method can be used as a customer identifier to query whether the customer information of the customer is stored on the CRM server.
- you arrive you can directly obtain the customer's information from the CRM server. Knowing the customer's needs before communicating again can help improve communication efficiency. It is also possible to update the customer's customer profile based on the re-communicated recording data.
- Embodiments of the present invention also provide a system comprising: one or more processors; a memory; one or more programs, the one or more programs being stored in the memory; the one or more The operations of the voice data processing method according to the above described are performed when the processor executes the one or more programs. Regarding the respective operations of the voice data processing method of the present invention, details are not described herein again.
- Embodiments of the present invention also provide a non-volatile computer storage medium storing one or more programs when the one or more programs are executed by one or more devices Having the one or more devices perform various operations in accordance with the voice data processing method described above. Regarding the respective operations of the voice data processing method of the present invention, details are not described herein again.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Telephonic Communication Services (AREA)
Abstract
本发明提供了一种语音数据处理方法和装置。该方法包括：响应第一用户发起的录音解析请求，根据所述录音解析请求中的客户标识获取与所述客户标识对应的客户的录音文件，将所述录音文件转换为文本文件，从转换后的文本文件中提取涉及所述客户的一个或多个关键字作为所述客户的客户资料，所述关键字包括所述客户的单位名称和关注信息。实施上述方法和装置，改善了客户资料数据获取的处理效率，方便了客户资料的完善和管理。
Description
本申请要求于2015年1月14日提交中国知识产权局、申请号为201510018632.5、发明名称为“一种语音数据处理方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
本发明涉及数据处理领域，更为具体而言，涉及一种语音数据处理方法和装置。
随着互联网的迅速普及和搜索技术的飞速发展，在企业进行业务推广时，通过互联网搜索辅助寻找潜在客户已经是一个广泛应用的方式。一般而言，在业务推广人员进行业务推广时，最常用的方式是通过电话以语音形式与潜在客户就业务推广的相关问题进行沟通。然而，这样的语音沟通方式，由于推广人员和潜在客户各自的不同的语言习惯、语言风格、语调、语速等不确定因素，导致语音形式的沟通录音数据的可用性较低，对于录音数据中涉及潜在客户的关键数据的保存和完善效率更低。因此，目前亟需改善录音数据可用性的解决方案。
发明内容
为解决上述技术问题，本发明提供了一种语音数据处理方法和装置，通过将推广人员与客户沟通的录音数据转换为文本数据，将录音数据的处理由复杂的听转换为便捷的看，提高了录音数据的可用性；并且基于转换后的文本数据提取客户沟通中的关键字作为该客户的客户资料，改善了客户资料数据获取的处理效率，方便了客户资料的完善和管理。
根据本发明实施方式的第一方面，提供了一种语音数据处理方法，该方法可包括：
响应第一用户发起的录音解析请求，根据所述录音解析请求中的客户标识获取与所述客户标识对应的客户的录音文件，
将所述录音文件转换为文本文件，从转换后的文本文件中提取涉及所述客户的一个或多个关键字作为所述客户的客户资料，所述关键字包括所述客户的单位名称和关注信息。
在本发明的一些实施方式中，所述方法可进一步包括：将提取的所述客户的单位名称与行政机关数据库中的登记名称进行比对以获取所述客户所在单位的登记名称作为所述客户资料的第一补充部分，其中所述行政机关数据库包括工商机关数据库、民政机关数据库中的一者或一者以上。
在本发明的一些实施方式中，所述方法可进一步包括：将提取的所述客户的单位名称与从获取所述客户沟通方式的网页中抓取的企业名称进行比对以获取所述客户所在单位的网页名称作为所述客户资料的第二补充部分。
在本发明的一些实施方式中，所述方法可进一步包括：将所述客户资料通过浏览器插件向以文本形式向所述第一用户显示，并将所述客户资料上传至服务器。
在本发明的一些实施方式中，所述方法可进一步包括：响应第二用户的客户资料解析请求，并根据所述客户资料解析请求中的客户标识向所述第二用户发送所述客户标识对应的客户资料，以便所述第二用户根据所述客户资料与所述客户再次沟通，并基于所述再次沟通更新所述客户资料。
根据本发明实施方式的第二方面，提供了一种语音数据处理装置，所述装置可包括：
数据交换模块，用于响应第一用户发起的录音解析请求，根据所述录音解析请求中的客户标识获取与所述客户标识对应的客户的录音文件，
语音识别模块，用于将所述录音文件转换为文本文件，从转换后的文本文件中提取涉及所述客户的一个或多个关键字作为所述客户的客户资料，所述关键字包括所述客户的单位名称和关注信息。
在本发明的一些实施方式中，所述语音识别模块，还可用于将提取的所述客户的单位名称与行政机关数据库中的登记名称进行比对以获取所述客户所在单位的登记名称作为所述客户资料的第一补充部分，其中，所述行政机关数据库包括工商机关数据库、民政机关数据库中的一者或一者以上。
在本发明的一些实施方式中，所述语音识别模块，还可用于将提取的所述客户的单位名称与从获取所述客户沟通方式的网页中抓取的企业名称进行比对
以获取所述客户所在单位的网页名称作为所述客户资料的第二补充部分。
在本发明的一些实施方式中，所述数据交换模块，还可用于将所述客户资料通过浏览器插件向以文本形式向所述第一用户显示，并将所述客户资料上传至服务器。
在本发明的一些实施方式中，所述服务器，还可用于响应第二用户的客户资料解析请求，并根据所述客户资料解析请求中的客户标识向所述第二用户发送所述客户标识对应的客户资料，以便所述第二用户根据所述客户资料与所述客户再次沟通，并基于所述再次沟通更新所述客户资料。
并且，本发明实施方式提供了一种系统，该系统可包括：一个或多个处理器；存储器；一个或多个程序，所述一个或多个程序存储于所述存储器；所述一个或多个处理器执行所述一个或多个程序时进行根据上述语音数据处理方法的操作。
本发明实施方式还提供了一种非易失性计算机存储介质，所述非易失性计算机存储介质存储有一个或多个程序，当所述一个或多个程序被一个或多个设备执行时，使得所述一个或多个设备执行根据上述语音数据处理方法的操作。
本发明实施方式提供的语音数据处理方法和装置，通过将与客户沟通的录音数据转换为文本数据，并基于转换后的文本数据提取客户的关键信息作为客户资料，提高了沟通中语音数据的信息利用率，改善了录音数据的可用性，从而方便了客户资料的完善和管理；其次，将基于转换后文本数据提取的单位名称与官方的数据库进行比对获取客户单位的准确登记名称和/或将基于转换后文本数据提取的单位名称与获取客户沟通方式的网页中抓取的企业名称进行比对获取客户单位的网页名称，进一步提高了录音数据中的单位信息的利用率，改善了客户资料的准确度和覆盖面；再次，将文本形式的客户资料通过浏览器插件向用户显示，改善了推广人员获取客户资料的便捷度，让推广人员能够实时或及时获取客户资料，了解客户需求，也有助于改善沟通效率；再次，通过将客户资料上传至服务器，便于推广人员获取其他推广人员在先沟通得到的客户资料，扩大了录音数据的使用人群，改善录音数据的信息共享程度，而且能够通过多个推广人员的多次沟通，及时更新客户资料。
图1概要图示了根据本发明实施方式的一种语音数据处理方法的流程示意图；
图2概要图示了根据本发明实施方式的一种语音数据处理装置的结构示意图。
为使本发明的实施例的目的、技术方案和优点更加清楚，下面将结合附图对本发明作进一步地详细描述。
参见图1，图示了根据本发明实施方式的一种语音数据处理方法的流程示意图，该方法具体可包括下述步骤：
步骤S101，响应第一用户发起的录音解析请求，根据录音解析请求中的客户标识获取与客户标识对应的客户的录音文件，
步骤S102，将录音文件转换为文本文件，从转换后的文本文件中提取涉及该客户的一个或多个关键字作为该客户的客户资料，其中，关键字包括该客户的单位名称和关注信息。
在本发明的实施方式中，语音数据处理方法可包括步骤S101响应第一用户发起的录音解析请求，根据录音解析请求中的客户标识获取与客户标识对应的客户的录音文件。其中，第一用户可以是推广人员，推广人员通过网页或其他方式获取客户的沟通方式(例如，电话号码、手机号码等)，并根据获取的沟通方式与客户进行语音沟通，并对语音沟通进行录音。
在一些应用场景下，步骤S101中，可响应推广人员通过客户端(例如，安装于推广人员计算设备上的客户端)插件发起与当前通话客户的实时的录音解析请求，该录音解析请求可包括客户标识(例如，可以是客户的电话号码或手机号码等)。在另外一些应用场景下，步骤S101中，可响应推广人员通过客户端插件发起在先的与某一客户的录音解析请求，例如，推广人员A与客户X进行了一次沟通并对沟通过程进行了录音，由于客户此时需求意向并不明朗，就未请求实时录音解析。一段时间之后，推广人员B也获取了客户X的沟通方式(例如，电话号码或手机号码等)，那么推广人员B在查询到录音文件服务器保存有该客户X的录音文件之后，可将客户X的沟通方式作为客户标识发起非实时的录音解析请求。
执行步骤S101，响应推广人员发起的实时/非实时的录音解析请求，并根据录音解析请求中的客户标识(例如，电话号码或手机号码)从存储录音文件的服务器获取(例如，通过下载方式)与该客户标识对应的客户的录音文件。接下来，执行步骤S102将所获取的录音文件转换为文本文件，例如，在转换的过程中可以使用深度神经网络技术学习录音数据的声学特征以对录音数据进行优化。随后，从转换后的文本文件中提取涉及该客户的一个或多个关键字作为该客户的客户资料，其中，一个或多个关键字可包括该客户的单位名称和关注信息，关注信息可以根据产品或客户等关联因素的不同而不同，例如，可包括但不限于：推广产品名称、产品型号、产品价格、产品产地、运输费用等。客户资料可包括从转换后的文本数据中提取的客户的单位名称、关注信息，客户资料除了包括上述内容之外，还可以包括推广人员与客户语音沟通的日期、通话时间等记录信息。录音数据由于各种不同的干扰因素(例如，通话背景杂音、通话双方的语音风格等)而使其中客户信息的可用性较低并且获取其中信息的手段比较耗时，本发明通过语音识别将录音数据转换为文本数据，并基于转换后的文本数据提取客户的关键信息作为客户资料，提高了沟通中语音数据的信息利用率和获取的便捷度，改善了录音数据的可用性，从而方便了客户资料的完善和管理。
在本发明的一些实施方式中，语音数据处理方法可进一步包括：将提取的该客户的单位名称与行政机关数据库中的登记名称进行比对以获取该客户所在单位的登记名称作为客户资料的第一补充部分，其中行政机关数据库包括工商机关数据库、民政机关数据库中的一者或一者以上。例如，将从转换后的文本数据中提取的客户单位名称与行政机关数据库中的登记名称进行比对，例如，对于一般的企业单位，则可以与工商机关数据库中的注册企业名称进行比对，对于一些社会团体、民办非企业单位等，则可以与民政机关数据库中的登记名称进行比对，从而获取该客户所在单位准确的登记名称，作为该客户资料的一个补充部分，该补充部分与客户资料的其他部分(例如，语音识别出的客户单位名称和关注信息等)一起保存于该客户的资料库中。本发明通过与底层大数据的上述对比可进一步提高录音数据中单位信息的利用率，改善客户资料的准确度。
在另外一些实施方式中，语音数据处理方法还可进一步包括：将提取的该
客户的单位名称与从获取该客户沟通方式的网页中抓取的企业名称进行比对以获取客户所在单位的网页名称作为该客户资料的第二补充部分。例如，可以将从转换后的文本数据中提取出的单位名称与搜索到该客户的沟通方式(例如，电话号码或手机号码等)的网页中抓取的企业名称进行比对，从而获取该客户的网页名称，例如，可以是该企业的简称或更容易引起社会关注的名称等，作为该客户的客户资料另一补充部分，该补充部分与客户资料的其他部分一起保存于该客户的资料库中。通过上述比对可进一步提高录音数据中单位信息的利用率，改善客户资料的覆盖面，便于客户资料的完善。
在本发明的一些实施方式中，语音数据处理方法可进一步包括：将客户资料通过浏览器插件向以文本形式向所述第一用户显示，并将客户资料上传至服务器。例如，将基于转换后的文本数据提取的文本形式的客户资料通过浏览器插件以浮层形式向推广人员显示，改善了录音数据使用的便捷度，也有助于提高推广的成功率。还可以将获取的客户资料上传至客户关系管理(Customer Relationship Management，CRM)服务器，便于客户资料管理和维护。
在本发明的一些实施方式中，语音数据处理方法可进一步包括：响应第二用户的客户资料解析请求，并根据客户资料解析请求中的客户标识向第二用户发送客户标识对应的客户资料，以便第二用户根据客户资料与该客户再次沟通，并基于再次沟通更新客户资料。其中，第二用户可以是与第一用户不同的另一用户(例如，另外的推广人员)。例如，推广人员从网页上搜索到某一客户的沟通方式(例如，电话号码、手机号码等)后，可以将该沟通方式作为客户标识查询CRM服务器上是否存储有该客户的客户资料，若查询到，则可以从CRM服务器直接获取该客户的资料，再次沟通之前了解该客户的需求，有助于提高沟通的效率。而且还可以基于再次沟通的录音数据，对该客户的客户资料进行更新。
以上结合具体实施例描述了本发明的语音数据处理方法，下面将结合具体实施例描述与上述处理方法对应的语音数据处理装置。
参见图2，概要图示了根据本发明实施方式的一种语音数据处理装置的结构示意图，语音数据处理装置200具体可包括：
数据交换模块201，用于响应第一用户发起的录音解析请求，根据所述录音解析请求中的客户标识获取与所述客户标识对应的客户的录音文件，
语音识别模块202，用于将所述录音文件转换为文本文件，从转换后的文本文件中提取涉及所述客户的一个或多个关键字作为所述客户的客户资料，所述关键字包括所述客户的单位名称和关注信息。
在本发明的实施方式中，语音数据处理装置200的各模块可分别设置于物理上分离但可通信的不同设备上，例如，数据交换模块201可设置于数据交换服务器，语音识别模块202可设置于语音服务器集群上，数据交换服务器和语音服务器集群可位于物理分离的位置，并且各模块可以与服务器设备上的其他模块相互调用。
本发明的数据交换模块201响应第一用户发起的录音解析请求，根据所述录音解析请求中的客户标识获取与所述客户标识对应的客户的录音文件。其中，第一用户可以是推广人员，推广人员通过网页或其他方式获取客户的沟通方式(例如，电话号码、手机号码等)，并根据获取的沟通方式与客户进行语音沟通，并对语音沟通进行录音。
在一些应用场景下，数据交换模块201可响应推广人员通过客户端(例如，安装于推广人员计算设备上的客户端)插件发起与当前通话客户的实时的录音解析请求，该录音解析请求可包括客户标识(例如，可以是客户的电话号码或手机号码等)。在另外一些应用场景下，数据交换模块201可响应推广人员通过客户端插件发起在先的与某一客户的非实时录音解析请求，例如，推广人员A与客户X进行了一次沟通并对沟通过程进行了录音，由于客户此时需求意向并不明朗，就未请求实时录音解析。一段时间之后，推广人员B也获取了客户X的沟通方式(例如，电话号码或手机号码等)，那么推广人员B在查询到录音文件服务器保存有该客户X的录音文件之后，可将客户X的沟通方式作为客户标识发起非实时的录音解析请求。
数据交换模块201可响应推广人员发起的实时/非实时的录音解析请求，并根据录音解析请求中的客户标识(例如，电话号码或手机号码)从存储录音文件的服务器获取(例如，通过下载方式)与该客户标识对应的客户的录音文件，然后将所获取的录音文件发送至语音识别模块202。语音识别模块202将所获取的录音文件转换为文本文件，例如，在转换的过程中可以使用深度神经网络技术学习录音数据的声学特征以对录音数据进行优化，并且从转换后的文本文件中提取涉及该客户的一个或多个关键字作为该客户的客户资料，其中，该一个或
多个关键字可包括该客户的单位名称和关注信息，关注信息可以根据产品或客户等关联因素的不同而不同，例如，可包括但不限于：推广产品名称、产品型号、产品价格、产品产地、运输费用等。客户资料可包括从转换后的文本数据中提取的客户的单位名称、关注信息，客户资料除了包括上述内容之外还可以包括推广人员与客户语音沟通的日期、通话时间等记录信息。录音数据由于各种不同的干扰因素(例如，通话背景杂音、通话双方的语音风格等)而使其中客户信息的可用性较低并且获取其中信息的手段比较耗时，本发明通过语音识别将录音数据转换为文本数据，并基于转换后的文本数据提取客户的关键信息作为客户资料，提高了沟通中语音数据的信息利用率和获取的便捷度，改善了录音数据的可用性，从而方便了客户资料的完善和管理。
在本发明的一些实施方式中，语音识别模块202还可用于将提取的该客户的单位名称与行政机关数据库中的登记名称进行比对以获取该客户所在单位的登记名称作为客户资料的第一补充部分，其中行政机关数据库包括工商机关数据库、民政机关数据库中的一者或一者以上。例如，将从转换后的文本数据中提取的客户单位名称与行政机关数据库中的登记名称进行比对，例如，对于一般的企业单位，则可以与工商机关数据库中的注册企业名称进行比对，对于一些社会团体、民办非企业单位等，则可以与民政机关数据库中的登记名称进行比对，从而获取该客户所在单位准确的登记名称，作为该客户资料的一个补充部分，该补充部分与客户资料的其他部分(例如，语音识别出的客户单位名称和关注信息等)一起保存于该客户的资料库中。本发明通过与底层大数据的上述对比可进一步提高录音数据中单位信息的利用率，改善客户资料的准确度。
在另外一些实施方式中，语音识别模块202还可用于将提取的该客户的单位名称与从获取该客户沟通方式的网页中抓取的企业名称进行比对以获取客户所在单位的网页名称作为该客户资料的第二补充部分。例如，可以将从转换后的文本数据中提取出的单位名称与搜索到该客户的沟通方式(例如，电话号码或手机号码等)的网页中抓取的企业名称进行比对，从而获取该客户的网页名称，例如，可以是该企业的简称或更容易引起社会关注的名称等，作为该客户的客户资料另一补充部分，该补充部分与客户资料的其他部分一起保存于该客户的资料库中。通过上述比对可进一步提高录音数据中单位信息的利用率，改善客户资料的覆盖面，便于客户资料的完善。
在本发明的一些实施方式中，数据交换模块201还可用于将客户资料通过浏览器插件向以文本形式向所述第一用户显示，并将客户资料上传至服务器。例如，将基于转换后的文本数据提取的文本形式的客户资料通过浏览器插件以浮层形式向推广人员显示，改善了录音数据使用的便捷度，也有助于提高推广的成功率。还可以将获取的客户资料上传至客户关系管理CRM服务器，便于客户资料管理和维护。
在本发明的一些实施方式中，接收数据交换模块201上传的客户资料的CRM服务器还可用于：响应第二用户的客户资料解析请求，并根据客户资料解析请求中的客户标识向第二用户发送客户标识对应的客户资料，以便第二用户根据客户资料与该客户再次沟通，并基于再次沟通更新客户资料。其中，第二用户可以是与第一用户不同的另一用户(例如，另外的推广人员)。例如，推广人员从网页上搜索到某一客户的沟通方式(例如，电话号码、手机号码等)后，可以将该沟通方式作为客户标识查询CRM服务器上是否存储有该客户的客户资料，若查询到，则可以从CRM服务器直接获取该客户的资料，再次沟通之前了解该客户的需求，有助于提高沟通的效率。而且还可以基于再次沟通的录音数据，对该客户的客户资料进行更新。
本发明实施方式还提供了一种系统，所述系统包括：一个或多个处理器；存储器；一个或多个程序，所述一个或多个程序存储于所述存储器；所述一个或多个处理器执行所述一个或多个程序时进行根据上面描述的语音数据处理方法的各个操作。关于本发明的语音数据处理方法的各个操作，在此不再赘述。
本发明实施方式还提供了一种非易失性计算机存储介质，所述非易失性计算机存储介质存储有一个或多个程序，当所述一个或多个程序被一个或多个设备执行时，使得所述一个或多个设备执行根据上面描述的语音数据处理方法的各个操作。关于本发明的语音数据处理方法的各个操作，在此不再赘述。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到本发明可借助软件结合硬件平台的方式来实现，当然也可以全部通过硬件来实施。基于这样的理解，本发明的技术方案对背景技术做出贡献的全部或者部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在存储介质中，如ROM/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，智能手机或者网络设备等)执行本发明各个实施例或者实
施例的某些部分所述的方法。
本发明说明书中使用的术语和措辞仅仅为了举例说明，并不意味构成限定。本领域技术人员应当理解，在不脱离所公开的实施方式的基本原理的前提下，对上述实施方式中的各细节可进行各种变化。因此，本发明的范围只由权利要求确定，在权利要求中，除非另有说明，所有的术语应按最宽泛合理的意思进行理解。
Claims (12)
- 一种语音数据处理方法，其特征在于，包括：响应第一用户发起的录音解析请求，根据所述录音解析请求中的客户标识获取与所述客户标识对应的客户的录音文件，将所述录音文件转换为文本文件，从转换后的文本文件中提取涉及所述客户的一个或多个关键字作为所述客户的客户资料，所述关键字包括所述客户的单位名称和关注信息。
- 根据权利要求1所述的方法，其特征在于，所述方法进一步包括：将提取的所述客户的单位名称与行政机关数据库中的登记名称进行比对以获取所述客户所在单位的登记名称作为所述客户资料的第一补充部分，其中，所述行政机关数据库包括工商机关数据库、民政机关数据库中的一者或一者以上。
- 根据权利要求2所述的方法，其特征在于，所述方法进一步包括：将提取的所述客户的单位名称与从获取所述客户沟通方式的网页中抓取的企业名称进行比对以获取所述客户所在单位的网页名称作为所述客户资料的第二补充部分。
- 根据权利要求1至3中任意一项所述的方法，其特征在于，所述方法进一步包括：将所述客户资料通过浏览器插件向以文本形式向所述第一用户显示，并将所述客户资料上传至服务器。
- 根据权利要求4所述的方法，其特征在于，所述方法进一步包括：响应第二用户的客户资料解析请求，并根据所述客户资料解析请求中的客户标识向所述第二用户发送所述客户标识对应的客户资料，以便所述第二用户根据所述客户资料与所述客户再次沟通，并基于所述再次沟通更新所述客户资料。
- 一种语音数据处理装置，其特征在于，包括：数据交换模块，用于响应第一用户发起的录音解析请求，根据所述录音解析请求中的客户标识获取与所述客户标识对应的客户的录音文件，语音识别模块，用于将所述录音文件转换为文本文件，从转换后的文本文件中提取涉及所述客户的一个或多个关键字作为所述客户的客户资料，所述关键字包括所述客户的单位名称和关注信息。
- 根据权利要求6所述的装置，其特征在于，所述语音识别模块，还用于将提取的所述客户的单位名称与行政机关数据库中的登记名称进行比对以获取所述客户所在单位的登记名称作为所述客户资料的第一补充部分，其中，所述行政机关数据库包括工商机关数据库、民政机关数据库中的一者或一者以上。
- 根据权利要求7所述的装置，其特征在于，所述语音识别模块，还用于将提取的所述客户的单位名称与从获取所述客户沟通方式的网页中抓取的企业名称进行比对以获取所述客户所在单位的网页名称作为所述客户资料的第二补充部分。
- 根据权利要求6至8中任意一项所述的装置，其特征在于，所述数据交换模块，还用于将所述客户资料通过浏览器插件向以文本形式向所述第一用户显示，并将所述客户资料上传至服务器。
- 根据权利要求9所述的装置，其特征在于，所述服务器，还用于响应第二用户的客户资料解析请求，并根据所述客户资料解析请求中的客户标识向所述第二用户发送所述客户标识对应的客户资料，以便所述第二用户根据所述客户资料与所述客户再次沟通，并基于所述再次沟通更新所述客户资料。
- 一种系统，其特征在于，所述系统包括：一个或多个处理器；存储器；一个或多个程序，所述一个或多个程序存储于所述存储器；所述一个或多个处理器执行所述一个或多个程序时进行根据权利要求1至5中任意一项所述的操作。
- 一种非易失性计算机存储介质，其特征在于，所述非易失性计算机存储介质存储有一个或多个程序，当所述一个或多个程序被一个或多个设备执行时，使得所述一个或多个设备执行根据权利要求1至5中任意一项所述的操作。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510018632.5 | 2015-01-14 | ||
| CN201510018632.5A CN104580477A (zh) | 2015-01-14 | 2015-01-14 | 一种语音数据处理方法和装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2016112665A1 true WO2016112665A1 (zh) | 2016-07-21 |
Family
ID=53095634
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2015/083943 WO2016112665A1 (zh) | 2015-01-14 | 2015-07-14 | 一种语音数据处理方法和装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN104580477A (zh) |
| WO (1) | WO2016112665A1 (zh) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109726372A (zh) * | 2017-10-31 | 2019-05-07 | 百度在线网络技术（北京）有限公司 | 基于通话记录的工单的生成方法、装置及计算机可读介质 |
| CN110310127A (zh) * | 2019-05-23 | 2019-10-08 | 中国平安人寿保险股份有限公司 | 录音获取方法、装置、计算机设备及存储介质 |
| CN111279304A (zh) * | 2017-09-29 | 2020-06-12 | 甲骨文国际公司 | 基于画布上连接的可定位元素配置通信决策树的方法和系统 |
| US11775843B2 (en) | 2017-09-29 | 2023-10-03 | Oracle International Corporation | Directed trajectories through communication decision tree using iterative artificial intelligence |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104580477A (zh) * | 2015-01-14 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | 一种语音数据处理方法和装置 |
| CN106933791A (zh) * | 2017-03-10 | 2017-07-07 | 山东浪潮云服务信息科技有限公司 | 一种生成证照方法及装置 |
| CN107172311B (zh) * | 2017-06-22 | 2020-06-19 | 深圳市佰仟金融服务有限公司 | 业务评估方法及终端设备 |
| CN107679874A (zh) * | 2017-09-28 | 2018-02-09 | 平安科技（深圳）有限公司 | 丰富微信客户标签的方法、电子装置及存储介质 |
| CN108549848B (zh) * | 2018-03-27 | 2022-02-25 | 百度在线网络技术（北京）有限公司 | 用于输出信息的方法和装置 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110013756A1 (en) * | 2009-07-15 | 2011-01-20 | Google Inc. | Highlighting of Voice Message Transcripts |
| CN102572372A (zh) * | 2011-12-28 | 2012-07-11 | 中兴通讯股份有限公司 | 会议纪要的提取方法和装置 |
| CN103354561A (zh) * | 2013-06-28 | 2013-10-16 | 贵阳朗玛信息技术股份有限公司 | 配置信息的推送方法、处理方法及装置 |
| CN103686377A (zh) * | 2013-12-09 | 2014-03-26 | 深圳Tcl新技术有限公司 | 推送资讯信息的方法和智能终端 |
| CN104580477A (zh) * | 2015-01-14 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | 一种语音数据处理方法和装置 |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1722757B (zh) * | 2004-07-12 | 2010-11-03 | 深圳市友邻通讯设备有限公司 | 基于话音通讯的录音系统 |
| CN1852354A (zh) * | 2005-10-17 | 2006-10-25 | 华为技术有限公司 | 收集用户行为特征的方法和装置 |
| CN101325626B (zh) * | 2007-06-15 | 2011-07-20 | 深圳市友邻通讯设备有限公司 | 基于网络语音通信的实时录音监察服务方法 |
-
2015
- 2015-01-14 CN CN201510018632.5A patent/CN104580477A/zh active Pending
- 2015-07-14 WO PCT/CN2015/083943 patent/WO2016112665A1/zh active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110013756A1 (en) * | 2009-07-15 | 2011-01-20 | Google Inc. | Highlighting of Voice Message Transcripts |
| CN102572372A (zh) * | 2011-12-28 | 2012-07-11 | 中兴通讯股份有限公司 | 会议纪要的提取方法和装置 |
| CN103354561A (zh) * | 2013-06-28 | 2013-10-16 | 贵阳朗玛信息技术股份有限公司 | 配置信息的推送方法、处理方法及装置 |
| CN103686377A (zh) * | 2013-12-09 | 2014-03-26 | 深圳Tcl新技术有限公司 | 推送资讯信息的方法和智能终端 |
| CN104580477A (zh) * | 2015-01-14 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | 一种语音数据处理方法和装置 |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111279304A (zh) * | 2017-09-29 | 2020-06-12 | 甲骨文国际公司 | 基于画布上连接的可定位元素配置通信决策树的方法和系统 |
| CN111279304B (zh) * | 2017-09-29 | 2023-08-15 | 甲骨文国际公司 | 基于画布上连接的可定位元素配置通信决策树的方法和系统 |
| US11775843B2 (en) | 2017-09-29 | 2023-10-03 | Oracle International Corporation | Directed trajectories through communication decision tree using iterative artificial intelligence |
| US11900267B2 (en) | 2017-09-29 | 2024-02-13 | Oracle International Corporation | Methods and systems for configuring communication decision trees based on connected positionable elements on canvas |
| CN109726372A (zh) * | 2017-10-31 | 2019-05-07 | 百度在线网络技术（北京）有限公司 | 基于通话记录的工单的生成方法、装置及计算机可读介质 |
| CN110310127A (zh) * | 2019-05-23 | 2019-10-08 | 中国平安人寿保险股份有限公司 | 录音获取方法、装置、计算机设备及存储介质 |
| CN110310127B (zh) * | 2019-05-23 | 2024-03-05 | 中国平安人寿保险股份有限公司 | 录音获取方法、装置、计算机设备及存储介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104580477A (zh) | 2015-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2016112665A1 (zh) | 一种语音数据处理方法和装置 | |
| US10997258B2 (en) | Bot networks | |
| US20210287250A1 (en) | Providing data and analysis for advertising on networked devices | |
| CN105027196B (zh) | 自动语音识别系统中的快速词汇表外搜索 | |
| CN105701254B (zh) | 一种信息处理方法和装置、一种用于信息处理的装置 | |
| US8903904B2 (en) | Pushing identity information | |
| US9275647B2 (en) | Periodic ambient waveform analysis for enhanced social functions | |
| CN103346957B (zh) | 一种根据联系人消息改变联系人头像表情的系统及方法 | |
| US20120109631A1 (en) | Providing multi-lingual translation for third party content feed applications | |
| CN105426508B (zh) | 网页生成方法和装置 | |
| US10462257B2 (en) | Method and apparatus for obtaining user account | |
| US9571443B2 (en) | Mobile device message enabled on-line community bulletin board | |
| CN113051389B (zh) | 知识推送方法及装置 | |
| US11973733B2 (en) | Enhancing a social media post with content that is relevant to the audience of the post | |
| JP6684894B2 (ja) | 情報をプッシュ配信する方法及び装置 | |
| US11599521B2 (en) | Systems and methods for providing real-time discrepancies between disparate execution platforms | |
| US20230153364A1 (en) | Information processing apparatus and method | |
| US10831845B2 (en) | Generation of enhanced search results | |
| US20160012495A1 (en) | Soliciting customer feedback based on indoor positioning system detection of physical customer presence | |
| US20220311766A1 (en) | Sensor-based authentication, notification, and assistance systems | |
| CN103176998A (zh) | 一种基于语音识别的阅读辅助系统 | |
| US20210056600A1 (en) | Systems and methods for real-time processing of audio feedback | |
| JP5823668B2 (ja) | 情報処理装置、メッセージ伝達方法およびプログラム | |
| TWI743160B (zh) | 業務資料處理方法及裝置 | |
| CN113515687B (zh) | 物流信息的获取方法和装置 |
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: 15877580 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: 15877580 Country of ref document: EP Kind code of ref document: A1 |