WO2016091110A1 - 语音信息处理方法及装置 - Google Patents

语音信息处理方法及装置 Download PDF

Info

Publication number
WO2016091110A1
WO2016091110A1 PCT/CN2015/096241 CN2015096241W WO2016091110A1 WO 2016091110 A1 WO2016091110 A1 WO 2016091110A1 CN 2015096241 W CN2015096241 W CN 2015096241W WO 2016091110 A1 WO2016091110 A1 WO 2016091110A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
voice data
priority
voice
channel
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2015/096241
Other languages
English (en)
French (fr)
Inventor
凌青
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
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 Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Priority to KR1020197008942A priority Critical patent/KR102284973B1/ko
Priority to KR1020177018385A priority patent/KR20170088997A/ko
Priority to EP15868151.0A priority patent/EP3232651B1/en
Priority to US15/532,934 priority patent/US10708423B2/en
Priority to SG11201704392QA priority patent/SG11201704392QA/en
Priority to JP2017530753A priority patent/JP6538846B2/ja
Publication of WO2016091110A1 publication Critical patent/WO2016091110A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
    • H04M3/5166—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing in combination with interactive voice response systems or voice portals, e.g. as front-ends
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
    • H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
    • H04M3/4936—Speech interaction details
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
    • H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
    • G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
    • G10L25/63—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for estimating an emotional state
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
    • H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
    • H04M3/4931—Directory assistance systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
    • H04M3/523—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing with call distribution or queueing
    • H04M3/5232—Call distribution algorithms
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/527—Centralised call answering arrangements not requiring operator intervention

Definitions

  • the present application relates to the field of voice processing technologies, and in particular, to a voice information processing method and apparatus.
  • IVR interactive voice response
  • the process of using the IVR includes: the user accesses the service center of the IVR by telephone, selects a service channel according to the voice prompt of the service center, and uses the selected service channel to conduct a problem consultation, and the selected service channel may be a manual service channel or Intelligent service channel.
  • the service model of IVR is relatively simple, and the service method is not flexible enough, resulting in low service quality of IVR.
  • aspects of the present application provide a voice information processing method and apparatus for improving the quality of service of an IVR.
  • An aspect of the present application provides a voice information processing method, including:
  • the user is connected to the corresponding service channel according to the priority of the user.
  • a voice information processing apparatus including:
  • An obtaining module configured to obtain voice data for describing a question that the user consults
  • An analysis module configured to perform sentiment analysis on the voice data to determine a priority of the user
  • a service access module configured to access the user according to the priority of the user Service channel.
  • the user's voice data is acquired, and the voice data is sentimentally analyzed to determine the priority of the voice data, and the user is connected to the corresponding service channel according to the priority of the voice data.
  • the application can access the corresponding service channel according to the priority of the voice data, so that the user can access the different service channels flexibly and timely, for example, the priority of the voice data can be compared.
  • High-level users access high-priority service channels, which helps improve user experience and problem-solving rate. At the same time, it expands the service mode of IVR, improves service flexibility, and helps improve IVR service quality.
  • FIG. 1 is a schematic flowchart of a voice information processing method according to an embodiment of the present disclosure
  • step 101 is a schematic flowchart of an implementation manner of step 101 according to an embodiment of the present application.
  • FIG. 3 is a schematic flowchart diagram of an implementation manner of step 102 according to an embodiment of the present application.
  • FIG. 4 is a schematic flowchart of a voice information processing method according to another embodiment of the present disclosure.
  • FIG. 5 is a schematic structural diagram of a voice information processing apparatus according to another embodiment of the present disclosure.
  • FIG. 1 is a schematic flowchart diagram of a voice information processing method according to an embodiment of the present application. As shown in Figure 1, the method includes:
  • This embodiment provides a voice information processing method, which can be executed by a voice information processing apparatus.
  • the voice information processing device can be any device or device with certain processing capabilities, and can be applied to a customer service system that provides voice services, such as an IVR.
  • the voice information processing device acquires the voice data of the user, performs sentiment analysis on the voice data, determines the priority of the voice data, and then accesses the user to the corresponding service channel according to the priority of the voice data.
  • the priority of voice data can indicate or reflect the priority of the user to a certain extent.
  • the sentiment analysis of the voice data determines the priority of the voice data, and the user can access the different service channels according to the priority of the voice data. For example, the user with higher priority of the voice data can be connected.
  • the service channel that enters the high priority does not select the service channel by the user according to the voice prompt as in the prior art, and expands the service mode of the IVR, thereby improving the service quality of the IVR.
  • the user's voice data may be voice data used to describe the questions the user is consulting.
  • various applications generally provide a customer service system for providing a communication platform for users who use the application, for users to consult questions, understand related information, and feedback.
  • the customer service system provides at least one service channel (or service mode), for example, may include a manual channel and an intelligent channel.
  • the manual channel mainly refers to a manner in which a service agent provides a service to a user by voice;
  • the intelligent channel is Refers to non-human service methods, such as online chat Day service and online self-service.
  • the service channel is prioritized.
  • the priority of the artificial channel is higher than the priority of the intelligent channel.
  • the artificial channel can be further divided into an ordinary artificial channel, an expert artificial channel, and the like.
  • the expert manual channel has a higher priority than the normal manual channel. The higher the priority of the service channel, the better the quality of service provided.
  • the expert human channel can provide higher quality of service than the normal artificial channel.
  • the quality of service of different service channels can be determined by various factors such as the difficulty of solving the problem, the attitude of the service, the speed of solving the problem, and the flexibility of solving the problem.
  • a specific process for applying the method provided by the embodiment in the customer service system includes: when a user needs a consultation question, acquiring voice data for describing a problem consulted by the user, performing sentiment analysis on the voice data, and obtaining a user.
  • the priority of the voice data is determined according to the emotion when the user consults the question, and then the user is connected to the corresponding service channel according to the priority of the voice data.
  • the priority of the voice data can be used to reflect the priority of the user, thereby realizing the distinction between the users, and then accessing different services to users of different priorities.
  • the channel is conducive to improving the user experience and problem solving rate, and is conducive to improving the service quality of the customer service system.
  • the method provided in this embodiment can reflect the priority of the user by the priority of the voice data, which is equivalent to distinguishing the user. For example, the user who is more urgent in the appeal can be preferentially connected to the artificial channel. It does not treat all users uniformly as in the prior art, and expands the service mode of the customer service system, thereby improving the flexibility of the service.
  • an implementation manner of step 101 is as shown in FIG. 2, and specifically includes:
  • the voice prompt tone may be a problem description prompt tone for indicating a user to describe the question to be consulted. Accordingly, the user can issue voice data for describing the question in question at the prompt of the question description prompt tone.
  • the customer service systems that provide voice services to users, such as IVR
  • they When the customer consults these customer service systems, they usually dial the corresponding customer service hotline through the communication terminal such as mobile phone and fixed line, and send a call request to the voice information processing device in the customer service system.
  • the voice information processing device receives the call request to establish with the user. Voice calls.
  • the voice information processing apparatus plays a question description prompt tone to the user to prompt the user to describe the question to be consulted.
  • the problem description tone can be similar to "please describe the question asked in one sentence", "what is the question to ask?"
  • the user describes the prompt tone according to the question, describes the question to be consulted, and forms voice data describing the question to be consulted.
  • the voice data may be "I forgot the password", "I can't log in normally after entering the account and password” and the like.
  • the voice information processing apparatus may adopt other methods in addition to the voice data for describing the question consulted by the user in the above manner.
  • the voice information processing apparatus may acquire voice data for describing a question consulted by the user from another device such as a user terminal.
  • step 102 is as shown in FIG. 3, and specifically includes:
  • 1011 Perform sentiment analysis on the voice data to determine the volume and/or the speech rate of the user when the voice data is sent.
  • sentiment analysis belongs to one of semantic analysis, and belongs to the prior art.
  • Various methods for analyzing volume and speech rate are applicable to the present embodiment.
  • the process of analyzing the speech rate may be: after acquiring the speech data, calculating the short-term average amplitude of the speech data, and then analyzing the start and end points of the speech data according to the short-term average amplitude, and calculating the speech rate of the speech data.
  • Sound is made up of several (or infinitely many) frequency components.
  • the process of analyzing the volume may be: converting the voice data from the time domain signal to the frequency domain signal, dividing the octave, and converting the frequency domain signal from the absolute magnitude to the relative magnitude. That is, decibel (DB), the volume of the voice data is obtained.
  • DB decibel
  • the voice information processing device obtains the speech rate and/or volume when the user describes the question consulted by performing sentiment analysis on the voice data.
  • the speech rate and volume can reflect the user's emotions, such as anxious, angry, calm, and so on.
  • the volume when the volume is between 0-20 decibels, it is very quiet and almost imperceptible; when the volume is between 20-40 decibels, it is quiet, like a soft whisper, indicating that the user's mood is relatively calm; when the volume is at Between 40-60 decibels, general, ordinary indoor conversation, indicating that the user's mood is relatively normal; when the volume is between 60-70 decibels, noisy, detrimental to the nerves, indicating that the user's mood is more anxious or angry.
  • the normal person's speaking speed per minute is 70-90 words in the general context. If the time for consulting the question in one sentence is 15 seconds, it should be 18-23. Words around. If the user's speaking speed is less than or equal to 18-23 words, it means that the user's emotions are relatively calm, and the question may be less urgent. If the user asks the question, the speech rate is greater than 23 words, for example. The speed of speech when consulting a question is about 25-32 words, which indicates that the user's mood is more irritable, and the question may be more urgent.
  • the voice information processing device obtains the user's emotion value based on the volume and/or the speech rate after obtaining the volume and/or speech rate of the user when describing the question in question.
  • the volume and/or the speech rate may be numerically processed, and the result of the numerical processing is taken as the emotional value of the user.
  • the method of numerical processing may be a weighted average or a weighted weight or the like.
  • an embodiment includes the voice information processing device using the product of the volume and the weight as the emotional value of the user.
  • the weight is 1, the volume can be directly used as the user's emotional value.
  • an embodiment includes the voice information processing apparatus using the product of the speech rate and the weight as the emotion value of the user.
  • the speech rate can be directly used as the user's emotional value.
  • an embodiment includes the voice information processing device performing numerical processing on the volume and the speech rate, and using the result of the numerical processing as the emotional value of the user. For example, the voice information processing device multiplies the volume and the speech rate, and uses the multiplied result as the emotion value of the user; or, the speech information processing device can perform weighted averaging on the volume and the speech rate, and use the result as the emotion value of the user. Numerical processing It is not limited to multiplication and weighted averaging.
  • the voice information processing device After obtaining the emotion value of the user, the voice information processing device matches the user's emotion value in the preset emotion value and the priority mapping relationship, and takes the priority in the matching as the priority of the user.
  • the relationship between the sentiment value and the priority mapping may be specifically: the correspondence between the range of sentiment values and the priority, for example, the sentiment value range 0-10 corresponds to the priority level-1, and the sentiment value range 10-20 corresponds to the priority level 0, The sentiment value range 20-30 corresponds to priority 1, and the sentiment value range of 30 or more corresponds to priority 2.
  • the relationship between the sentiment value and the priority mapping may also be: the correspondence between the sentiment value threshold and the priority, for example, the sentiment value threshold 10 corresponds to the priority level-1, the sentiment value threshold 20 corresponds to the priority level 0, and the sentiment value threshold value 30 corresponds to the priority. Level 1, the threshold value of the emotion value corresponding to priority 2 is not limited.
  • the voice information processing device presets a mapping relationship between the priority and the service channel, and the service channel includes an artificial channel and an intelligent channel. Further optionally, the artificial channel can be further divided into an ordinary artificial channel and an expert artificial channel.
  • the voice information processing device specifically matches the priority of the user in the mapping relationship between the preset priority and the service channel; when matching the artificial channel, the user is connected to the manual channel, specifically, the user's call request is transferred to On the client side of a customer service person in the manual channel.
  • the voice information processing device can guide the user to solve the problem by means of voice broadcast.
  • priority 2 corresponds to an expert manual channel
  • priority 1 corresponds to an ordinary artificial channel
  • priority 0 and priority-1 correspond to an intelligent channel. Assuming that the user's priority is 2, the voice information processing device needs to access the user to the expert manual channel.
  • the voice information processing device may further generate reference information for answering the question consulted by the user according to the voice data, and provide the reference information to the customer service personnel corresponding to the manual channel, so that the customer service personnel can solve the problem according to the reference information and improve the problem. Answer the efficiency and quality of the problem.
  • the manner of generating the reference information includes: but is not limited to: the semantic information processing device performs text conversion on the voice data, identifies the text information, determines a category to which the user consults the problem, and obtains the account information of the user, and the problem that the user consults belongs to.
  • the category and the user's account information are dynamically assembled to generate reference information.
  • the reference information can be in the form of a voice or a text. of.
  • the embodiment of the present application is not limited to determining the service channel that the user should access by using the priority of the user separately.
  • User priorities can also be combined with other information.
  • the priority of the user may be combined with a method for determining a user access service channel that has been implemented in the existing IVR (for example, determining a service channel that the user should access based on the category to which the user consults the question belongs).
  • the specific combination manner may include: prioritizing the user's priority, and then considering the category to which the user's consulting question belongs; or, prioritizing the category to which the user consults the question, and then considering the user's priority; or considering the user's consultation.
  • FIG. 4 provides a detailed flow for determining the service channel for the user using the priority of the user.
  • Each module in the embodiment shown in FIG. 4 may specifically constitute the above-described voice information processing apparatus.
  • FIG. 4 is a schematic flowchart diagram of a voice information processing method according to an embodiment of the present application. As shown in FIG. 4, the method includes:
  • the Interactive Voice Response (IVR) module receives the call request sent by the user to establish a voice call with the user.
  • the IVR module plays a problem description prompt tone to the user to prompt the user to describe the question to be consulted.
  • the IVR module receives voice data generated by the user's description of the problem being consulted.
  • the IVR module sends the voice data to the Alipay General Algorithm Platform (AGAP) module.
  • AGAP General Algorithm Platform
  • the AGAP module performs sentiment analysis on the voice data to obtain the user's volume and/or speech rate, and obtains the user's priority according to the volume and/or the speech rate, and returns the user's priority to the IVR module.
  • the IVR module sends the voice data to the Automatic Speech Recognition (ASR) module.
  • ASR Automatic Speech Recognition
  • the ASR module performs text conversion on the voice data, obtains text information, and returns the text information to the IVR module.
  • the IVR module provides the text information and the user's priority to the Customer Service Interactive Voice Response (CSIVR) module.
  • CSIVR Customer Service Interactive Voice Response
  • the CSIVR module sends the text information to the Alipay Text Analyze (ALITA) module.
  • ALITA Alipay Text Analyze
  • the ALITA module performs semantic analysis on the text information, determines the problem category, and returns the determined problem category to the CSIVR module.
  • the CSIVR module provides the user's caller ID and ID number to the Customer Information File (CIF) module.
  • CIF Customer Information File
  • the CIF module obtains the account information of the user according to the caller ID and the ID card number of the user, and returns the account information of the user to the CSIVR module.
  • the CSIVR module sends the user's account information, the problem category, and the user's priority to the Customer Service Accurate (CSACCURATE) module.
  • CSACCURATE Customer Service Accurate
  • the CSACCURATE module determines the service channel corresponding to the user according to the account information of the user, the problem category, and the priority of the user, and returns the identification information of the determined service channel to the CSIVR module.
  • the CSACCURATE module considers both the problem category and the user's priority to determine the service channel that the user should access.
  • the user belongs to the invalid user according to the category to which the question asked by the user belongs, and the priority of the user is determined to be -1 according to the volume and the speech rate when the user consults the question, and the user can be connected to the smart channel. It is assumed that the user belongs to the consulting class according to the category to which the user consults the question, and the priority of the user is determined to be 0 according to the volume and the speed of the question when the user consults the question, and the user can be connected to the smart channel.
  • the user belongs to the business appeal type user according to the category to which the user consults the question, and the user's priority is determined according to the volume and the speed of the user's consultation question, and the user can be connected to the normal artificial channel.
  • the category according to the question the user is consulting belongs to is determined that the user belongs to the emergency appeal type user, and the priority of the user is determined to be 2 according to the volume and the speech rate when the user consults the question, and the user can be connected to the expert artificial channel.
  • the CSIVR module If the service channel is an intelligent channel, the CSIVR module generates corresponding voice play data according to the user's account information, the problem category, and the user's priority, and returns the data to the IVR module for the IVR module to guide the user to solve the problem.
  • the CSIVR module If the service channel is an artificial channel, the CSIVR module generates reference information according to the user's account information, the problem category, and the user's priority, and connects the user to the artificial channel, and provides the reference information to the customer service personnel in the manual channel.
  • the priority of the user may be determined according to the emotion when the user consults the question, and the user is accessed to the corresponding service channel based on the priority, so that users of different priorities can be flexibly and timely accessed to different services.
  • Channels for example, can access higher priority users to higher priority service channels, which helps improve user experience and problem resolution rate.
  • it expands the service mode of IVR, improves service flexibility, and helps improve IVR. service quality.
  • FIG. 5 is a schematic structural diagram of a voice information processing apparatus according to an embodiment of the present disclosure. As shown in FIG. 5, the apparatus includes an acquisition module 51, an analysis module 52, and a service access module 53.
  • the obtaining module 51 is configured to acquire voice data of the user.
  • the analysis module 52 is connected to the acquisition module 51 for performing sentiment analysis on the voice data acquired by the acquisition module 51 to determine the priority of the voice data.
  • the service access module 53 is connected to the analysis module 52 for accessing the corresponding service channel according to the priority of the voice data determined by the analysis module 52.
  • the priority of voice data can indicate or reflect the priority of the user to a certain extent.
  • the user's voice data may be voice data used to describe the questions the user is consulting.
  • the obtaining module 51 is specifically configured to: receive a call request of the user to perform a voice call with the user; play a voice prompt tone to the user; and receive voice data sent by the user under the prompt of the voice prompt tone.
  • the voice prompt tone may be a problem description prompt tone for indicating a user to describe the question to be consulted. Accordingly, the user can issue voice data for describing the question in question at the prompt of the question description prompt tone.
  • the analyzing module 52 is specifically configured to: perform sentiment analysis on the voice data, determine a volume and/or a speech rate of the user when the voice data is sent; and obtain the user's emotional value according to the volume and/or the speech rate.
  • the priority of the voice data is obtained according to the user's emotional value and the preset emotional value and priority mapping relationship.
  • the analysis module 52 is specifically configured to: perform sentiment analysis on the voice data, determine the volume and/or the speech rate of the user when the voice data is sent; perform numerical processing on the volume and/or the speech rate, and use the result of the numerical processing as the emotional value of the user.
  • the priority of the voice data is obtained according to the user's emotional value and the preset emotional value and priority mapping relationship.
  • the service access module is specifically configured to: match a priority of the voice data in a mapping relationship between the preset priority and the service channel; and when the match is matched to the artificial channel, access the user to the manual aisle.
  • the voice information processing apparatus further includes: a reference processing module, configured to generate reference information according to the voice data, and provide the reference information to the customer service personnel corresponding to the manual channel.
  • a reference processing module configured to generate reference information according to the voice data, and provide the reference information to the customer service personnel corresponding to the manual channel.
  • the voice information processing apparatus may also adopt an implementation structure in the embodiment shown in FIG. 2 to form an IVR system.
  • the voice information processing apparatus may be used according to the priority of the voice data.
  • the user accesses the corresponding service channel, so that the user can access the different service channels in a flexible and timely manner. For example, a user with a higher priority of voice data can access the service channel with a higher priority, which is beneficial to improving the user experience and The problem solving rate, while expanding the service mode of the IVR, improves the flexibility of the service and helps improve the service quality of the IVR.
  • the disclosed system, apparatus, and method may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.
  • the above-described integrated unit implemented in the form of a software functional unit can be stored in a computer readable storage medium.
  • the software functional unit described above is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to perform the methods described in various embodiments of the present application. Part of the steps.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), A variety of media that can store program code, such as random access memory (RAM), disk, or optical disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Human Computer Interaction (AREA)
  • Health & Medical Sciences (AREA)
  • Hospice & Palliative Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Child & Adolescent Psychology (AREA)
  • Psychiatry (AREA)
  • Computational Linguistics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Telephonic Communication Services (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本申请提供一种语音信息处理方法及装置。方法包括：获取用户的语音数据；对所述语音数据进行情感分析，以确定语音数据的优先级；根据语音数据的优先级，将所述用户接入对应的服务通道。本申请可以提高IVR的服务质量。

Description

语音信息处理方法及装置 【技术领域】
本申请涉及语音处理技术领域，尤其涉及一种语音信息处理方法及装置。
【背景技术】
随着互联网和通信技术的发展，用户可以通过网络或电话进行通信，在通信过程中咨询一些问题并获得相关答案。例如，越来越多的客服系统采用互动式语音应答热线服务(Interactive Voice Response，IVR)方式，与用户进行通信，以便用户咨询一些问题，并向用户提供相关答案。
使用IVR的过程包括：用户通过电话接入IVR的服务中心，根据服务中心的语音提示选择一种服务通道，并使用所选择的服务通道进行问题咨询，所选择的服务通道可以是人工服务通道或智能服务通道。目前，IVR的服务模式比较单一，服务方式不够灵活，导致IVR的服务质量较低。
【发明内容】
本申请的多个方面提供一种语音信息处理方法及装置，用以提高IVR的服务质量。
本申请的一方面，提供一种语音信息处理方法，包括：
获取用于描述用户所咨询的问题的语音数据；
对所述语音数据进行情感分析，以确定所述用户的优先级；
根据所述用户的优先级，将所述用户接入对应的服务通道。
本申请的另一方面，提供一种语音信息处理装置，包括：
获取模块，用于获取用于描述用户所咨询的问题的语音数据；
分析模块，用于对所述语音数据进行情感分析，以确定所述用户的优先级；
服务接入模块，用于根据所述用户的优先级，将所述用户接入对应的服 务通道。
在本申请中，获取用户的语音数据，对该语音数据进行情感分析，以确定语音数据的优先级，根据语音数据的优先级，将用户接入对应的服务通道。与现有技术相比，本申请可以根据语音数据的优先级，将用户接入对应的服务通道，这样就可以灵活及时的将用户接入不同的服务通道，例如可以将语音数据的优先级较高的用户接入高优先级的服务通道，有利于提高用户体验和问题解决速率，同时扩展了IVR的服务模式，提高了服务的灵活性，有利于提高IVR的服务质量。
【附图说明】
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1为本申请一实施例提供的语音信息处理方法的流程示意图；
图2为本申请一实施例提供的步骤101的实施方式的流程示意图；
图3为本申请一实施例提供的步骤102的实施方式的流程示意图；
图4为本申请另一实施例提供的语音信息处理方法的流程示意图；
图5为本申请另一实施例提供的语音信息处理装置的结构示意图。
【具体实施方式】
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获 得的所有其他实施例，都属于本申请保护的范围。
图1为本申请一实施例提供的语音信息处理方法的流程示意图。如图1所示，该方法包括：
101、获取用户的语音数据。
102、对上述语音数据进行情感分析，以确定所述语音数据的优先级。
103、根据上述语音数据的优先级，将用户接入对应的服务通道。
本实施例提供一种语音信息处理方法，可由语音信息处理装置来执行。在具体实现上，语音信息处理装置可以是任何具有一定处理能力的装置或设备，并可应用于提供语音服务的客服系统，如IVR中。
具体的，语音信息处理装置获取用户的语音数据，对语音数据进行情感分析，以确定语音数据的优先级，进而根据语音数据的优先级，将用户接入对应的服务通道。
值得说明的是，语音数据的优先级一定程度上可以表示或反映用户的优先级。
本实施例通过对语音数据进行情感分析，确定语音数据的优先级，可以根据语音数据的优先级灵活及时的将用户接入不同的服务通道，例如可以将语音数据的优先级较高的用户接入高优先级的服务通道，不再像现有技术那样根据语音提示由用户选择服务通道，扩展IVR的服务模式，提高了IVR的服务质量。
在一可选实施方式中，用户的语音数据可以是用于描述用户所咨询的问题的语音数据。
在实际应用中，各种应用一般都会提供客服系统，用于向使用该应用的用户提供一种交流平台，以供用户进行问题咨询、相关信息的了解、以及意见反馈等。
在本实施例中，客服系统提供至少一种服务通道(或服务方式)，例如可以包括人工通道和智能通道，人工通道主要是指由客服人员以语音方式向用户提供服务的方式；智能通道是指非人工的服务方式，例如可包括在线聊 天服务和在线自助服务等。
在本实施例中，对服务通道进行优先级划分。人工通道的优先级要高于智能通道的优先级。另外，人工通道进一步可以分为普通人工通道、专家人工通道等。专家人工通道的优先级高于普通人工通道的优先级。优先级越高的服务通道所提供的服务质量越好，例如专家人工通道所能提供的服务质量要高于普通人工通道。不同服务通道的服务质量可由所能解决问题的难以程度、服务态度、解决问题的速度以及解决问题的灵活性等多方面因素来确定。
一种在客服系统中应用本实施例提供的方法的具体过程包括：当有用户需要咨询问题时，获取用于描述该用户所咨询的问题的语音数据，对该语音数据进行情感分析，获得用户在描述所咨询的问题时的情感，根据用户咨询问题时的情感确定语音数据的优先级，进而根据语音数据的优先级，将用户接入对应的服务通道。
由此可见，本实施例通过对用户咨询问题的语音数据进行情感分析，可通过语音数据的优先级体现用户的优先级，实现对用户的区分，进而将不同优先级的用户接入不同的服务渠道，有利于提高用户体验和问题解决速率，有利于提高客服系统的服务质量。另外，与现有技术相比，本实施例提供的方法可以通过语音数据的优先级体现用户的优先级，相当于对用户进行了区分，例如可以将诉求比较迫切的用户优先接入人工通道，不再像现有技术那样统一对待所有用户，扩展了客服系统的服务模式，提高了服务的灵活性。
在一可选实施方式中，步骤101的一种实现方式如图2所示，具体包括：
1011、接收用户的通话请求，以与用户进行语音通话。
1012、向用户播放语音提示音。
1013、接收用户在语音提示音的提示下发出的语音数据。
在一可选实施方式中，上述语音提示音可以是问题描述提示音，用于指示用户对所咨询的问题进行描述。相应的，用户可以在问题描述提示音的提示下发出用于描述所咨询的问题的语音数据。
详细来说，对于一些向用户提供语音服务的客服系统，例如IVR，当用 户向这些客服系统咨询问题时，一般会通过手机、固话等通信终端拨打相应的客服热线，向客服系统中的语音信息处理装置发送通话请求，语音信息处理装置接收通话请求，以与用户建立语音通话。之后，语音信息处理装置向用户播放问题描述提示音，以提示用户对所咨询的问题进行描述。例如，问题描述提示音可以是“请用一句话描述所咨询的问题”、“请问所要咨询的问题是什么”等类似语音。用户根据问题描述提示音，对所咨询的问题进行描述，形成描述所咨询的问题的语音数据。例如，语音数据可以是“我忘记密码了”、“输入账号和密码后无法正常登录”等类似语音。
值得说明的是，语音信息处理装置除了可以采用上述方式获得用于描述用户所咨询的问题的语音数据之外，还可以采用其他方式。例如，语音信息处理装置可以从其他设备(例如用户终端)上获取用于描述用户所咨询的问题的语音数据。
当语音信息处理装置获取到语音数据后，可以对语音数据进行情感分析，以确定语音数据的优先级。在一可选实施方式中，步骤102的一种实现方式如图3所示，具体包括：
1021、对语音数据进行情感分析，确定用户在发出语音数据时的音量和/或语速。
1022、根据上述音量和/或语速，获得用户的情感值。
1023、根据用户的情感值和预设的情感值与优先级映射关系，获得用户的优先级。
其中，情感分析属于语义分析中的一种，属于现有技术，各种可以分析出音量和语速的方法均适用于本实施例。举例说明，对语速的分析过程可以是：在获取到语音数据后，计算语音数据的短时平均振幅，然后根据短时平均振幅分析语音数据的起止点，计算出语音数据的语速。
声音是由若干(或无限多个)频率分量组成的。为了检测和描述声音的特性，一般使用频谱。因此对音量的分析过程可以是：把语音数据由时域信号转换为频域信号，划分倍频程，并把频域信号从绝对量值转换为相对量值， 即分贝(DB)，获得语音数据的音量。
语音信息处理装置通过对语音数据进行情感分析，获得用户描述所咨询的问题时的语速和/或音量。语速和音量可以反映用户的情感，例如着急、生气、平静等。
对于音量来说，当音量在0－20分贝之间时，很静、几乎感觉不到；当音量在20－40分贝之间时，安静、犹如轻声絮语，说明用户情绪比较平和；当音量在40－60分贝之间时，一般、普通室内谈话，说明用户情绪比较正常；当音量在60－70分贝之间时，吵闹、有损神经，说明用户情绪比较着急或生气。
对于语速来说，正常人在一般语境中每分钟的语速是70-90个字，若按照用一句话描述所咨询问题的时间是15秒钟计算，则应该在18个-23个字左右。若用户在咨询问题时的语速小于或在18-23个字左右，说明用户情感较为平和，可能所咨询问题不太着急；反之，若用户在咨询问题时的语速大于23个字，例如咨询问题时的语速在25个-32个字左右，说明用户情绪较为急躁，可能所咨询问题比较迫切。
基于此，语音信息处理装置在获得用户在描述所咨询的问题时的音量和/或语速之后，进一步根据音量和/或语速，获得用户的情感值。
可选的，可以对音量和/或语速进行数值处理，将数值处理的结果作为用户的情感值。数值处理的方式可以是加权平均或加权重等。
例如，一种实施方式包括：语音信息处理装置将音量与权重的乘积作为用户的情感值。当权重为1时，即可以直接将音量作为用户的情感值。
又例如，一种实施方式包括：语音信息处理装置将语速与权重的乘积作为用户的情感值。当权重为1时，即可以直接将语速作为用户的情感值。
又例如，一种实施方式包括：语音信息处理装置对音量和语速进行数值处理，将数值处理的结果作为用户的情感值。例如，语音信息处理装置将音量和语速相乘，将相乘的结果作为用户的情感值；或者，语音信息处理装置可以对音量和语速进行加权平均，将结果作为用户的情感值。数值处理方式 并不限于相乘和加权平均。
得到用户的情感值之后，语音信息处理装置将用户的情感值在预设的情感值与优先级映射关系中进行匹配，将匹配中的优先级作为用户的优先级。
值得说明的是，情感值与优先级映射关系具体可以是：情感值范围与优先级的对应关系，例如情感值范围0-10对应优先级-1，情感值范围10-20对应优先级0，情感值范围20-30对应优先级1,情感值范围30以上对应优先级2。或者，情感值与优先级映射关系具体也可以是：情感值阈值与优先级的对应关系，例如情感值阈值10对应优先级-1，情感值阈值20对应优先级0，情感值阈值30对应优先级1，关于优先级2对应的情感值阈值不做限定。
在一可选实施方式中，语音信息处理装置预先设定了优先级与服务通道的映射关系，且服务通道包括人工通道和智能通道。进一步可选的，人工通道又可分为普通人工通道和专家人工通道。语音信息处理装置具体将用户的优先级在预设的优先级与服务通道的映射关系中进行匹配；当匹配到人工通道时，将用户接入人工通道，具体是将用户的呼叫请求转接到人工通道中某个客服人员的客户端上。当匹配到智能通道时，将用户接入智能通道，具体可由语音信息处理装置以语音播报的方式指导用户解决问题。
举例说明，优先级2对应专家人工通道；优先级1对应普通人工通道；优先级0和优先级-1对应智能通道。假设用户的优先级为2，则语音信息处理装置需要将用户接入专家人工通道。
进一步，语音信息处理装置还可以根据语音数据，生成解答用户所咨询的问题的参考信息，将该参考信息提供给人工通道对应的客服人员，以便于客服人员根据该参考信息向用户解答问题，提高解答问题的效率和质量。
上述生成参考信息的方式包括但不限于：语义信息处理装置对语音数据进行文本转换，对文本信息进行识别确定用户所咨询的问题所属的类别，并获取用户的账户信息，将用户所咨询问题所属的类别以及用户的账户信息进行动态组装生成参考信息。该参考信息可以语音形式的，也可以是文本形式 的。
值得说明的是，虽然上述实施例重点说明了如何使用用户的优先级确定用户应该接入的服务通道，但本申请实施例并不限于单独使用用户的优先级确定用户应该接入的服务通道，用户的优先级还可以与其他信息相结合使用。举例说明，用户的优先级可以与现有IVR中已经实现的用于确定用户接入服务通道的方法(例如基于用户所咨询问题所属的类别确定用户应该接入的服务通道)相结合。具体结合方式可以包括：优先考虑用户的优先级，再考虑用户所咨询问题所属的类别；或者，可以优先考虑用户所咨询问题所属的类别，再考虑用户的优先级；或者，同时考虑用户所咨询问题所属的类别和用户的优先级。
下面图4所示实施例提供一种使用用户的优先级为用户确定服务通道的详细流程。图4所示实施例中的各模块具体可构成上述语音信息处理装置。
图4为本申请一实施例提供的语音信息处理方法的流程示意图。如图4所示，该方法包括：
21、互动式语音应答(Interactive Voice Response，IVR)模块接收用户发送的通话请求，以与用户建立语音通话。
22、IVR模块向用户播放问题描述提示音，以提示用户对所咨询的问题进行描述。
23、IVR模块接收用户描述所咨询的问题产生的语音数据。
24、IVR模块将语音数据发送给支付宝算法平台(Alipay General Algorithm Platform，AGAP)模块。
25、AGAP模块对语音数据进行情感分析，获得用户的音量和/或语速，并根据音量和/或语速获得用户的优先级，并将用户的优先级返回给IVR模块。
26、IVR模块将语音数据发送给自动语音识别技术(Automatic Speech Recognition，ASR)模块。
27、ASR模块对语音数据进行文本转换，获得文本信息，并将文本信息返回给IVR模块。
28、IVR模块将文本信息和用户的优先级提供给客服语音交互响应(Customer Service Interactive Voice Response，CSIVR)模块。
29、CSIVR模块将文本信息发送给文本分析平台(Alipay Text Analyze，ALITA)模块。
30、ALITA模块对文本信息进行语义分析，确定问题类别，并将所确定的问题类别返回给CSIVR模块。
31、CSIVR模块将用户的来电号码、身份证号等信息提供给用户信息文件(Customer Information File，CIF)模块。
32、CIF模块根据用户的来电号码、身份证号码获取用户的账号信息，并将用户的账号信息返回给CSIVR模块。
33、CSIVR模块将用户的账户信息、问题类别和用户的优先级发送给精准服务平台(Customer Service Accurate，CSACCURATE)模块。
34、CSACCURATE模块根据用户的账户信息、问题类别和用户的优先级，确定用户对应的服务通道，并将所确定的服务通道的标识信息返回给CSIVR模块。
在该实施方式中，CSACCURATE模块同时考虑问题类别和用户的优先级，来决定用户应该接入的服务通道。
举例说明，假设根据用户所咨询问题所属的类别判定用户属于无效用户，且根据用户咨询问题时的音量和语速确定用户的优先级为-1，则可以将用户接入智能通道。假设根据用户所咨询问题所属的类别判定用户属于咨询类用户，且根据用户咨询问题时的音量和语速确定用户的优先级为0，则可以将用户接入智能通道。假设根据用户所咨询问题所属的类别判定用户属于业务诉求类用户，且根据用户咨询问题时的音量和语速确定用户的优先级为1，则可以将用户接入普通人工通道。假设根据用户所咨询问题所属的类别 判定用户属于紧急诉求类用户，且根据用户咨询问题时的音量和语速确定用户的优先级为2，则可以将用户接入专家人工通道。
35、若服务通道是智能通道，则CSIVR模块根据用户的账户信息、问题类别和用户的优先级生成对应的语音播放资料，返回给IVR模块，以供IVR模块指导用户解决问题。
36、若服务通道是人工通道，则CSIVR模块根据用户的账户信息、问题类别和用户的优先级生成参考信息，将用户接入人工通道，同时将参考信息提供给人工通道中的客服人员。
在本实施例中，可以根据用户咨询问题时的情感，确定用户的优先级，基于优先级将用户接入对应的服务通道，这样就可以灵活及时的将不同优先级的用户接入不同的服务通道，例如可以将优先级较高的用户接入高优先级的服务通道，有利于提高用户体验和问题解决速率，同时扩展了IVR的服务模式，提高了服务的灵活性，有利于提高IVR的服务质量。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本申请并不受所描述的动作顺序的限制，因为依据本申请，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本申请所必须的。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。
图5为本申请一实施例提供的语音信息处理装置的结构示意图。如图5所示，该装置包括：获取模块51、分析模块52和服务接入模块53。
获取模块51，用于获取用户的语音数据。
分析模块52，与获取模块51连接，用于对获取模块51获取的语音数据进行情感分析，以确定语音数据的优先级。
服务接入模块53，与分析模块52连接，用于根据分析模块52确定的语音数据的优先级，将用户接入对应的服务通道。
值得说明的是，语音数据的优先级一定程度上可以表示或反映用户的优先级。
在一可选实施方式中，用户的语音数据可以是用于描述用户所咨询的问题的语音数据。
在一可选实施方式中，获取模块51具体用于：接收用户的通话请求，以与用户进行语音通话；向用户播放语音提示音；接收用户在语音提示音的提示下发出的语音数据。
在一可选实施方式中，上述语音提示音可以是问题描述提示音，用于指示用户对所咨询的问题进行描述。相应的，用户可以在问题描述提示音的提示下发出用于描述所咨询的问题的语音数据。
在一可选实施方式中，分析模块52具体用于：对语音数据进行情感分析，确定用户在发出语音数据时的音量和/或语速；根据音量和/或语速，获得用户的情感值；根据用户的情感值和预设的情感值与优先级映射关系，获得语音数据的优先级。
分析模块52具体用于：对语音数据进行情感分析，确定用户在发出语音数据时的音量和/或语速；对音量和/或语速进行数值处理，将数值处理的结果作为用户的情感值；根据用户的情感值和预设的情感值与优先级映射关系，获得语音数据的优先级。
在一可选实施方式中，服务接入模块具体用于：将语音数据的优先级在预设的优先级与服务通道的映射关系中进行匹配；当匹配到人工通道时，将用户接入人工通道。
在一可选实施方式中，该语音信息处理装置还包括：参考处理模块，用于根据语音数据，生成参考信息，并将参考信息提供给人工通道对应的客服人员。
值得说明的是，语音信息处理装置除了可以采用图5所示实现结构之外，还可以采用图2所示实施例中的实现结构，形成一个IVR系统。
本实施例提供的语音信息处理装置，可以根据语音数据的优先级，将用 户接入对应的服务通道，这样就可以灵活及时的将用户接入不同的服务通道，例如可以将语音数据的优先级较高的用户接入高优先级的服务通道，有利于提高用户体验和问题解决速率，同时扩展了IVR的服务模式，提高了服务的灵活性，有利于提高IVR的服务质量。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统，装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统，装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。
上述以软件功能单元的形式实现的集成的单元，可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介质中，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，或者网络设备等)或处理器(processor)执行本申请各个实施例所述方法的部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、 随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是：以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围。

Claims (12)

  1. 一种语音信息处理方法，其特征在于，包括：
    获取用户的语音数据；
    对所述语音数据进行情感分析，以确定所述语音数据的优先级；
    根据所述语音数据的优先级，将所述用户接入对应的服务通道。
  2. 根据权利要求1所述的方法，其特征在于，所述获取用户的语音数据，包括：
    接收所述用户的通话请求，以与所述用户进行语音通话；
    向所述用户播放语音提示音；
    接收所述用户在所述语音提示音的提示下发出的所述语音数据。
  3. 根据权利要求1所述的方法，其特征在于，所述对所述语音数据进行情感分析，以确定所述语音数据的优先级，包括：
    对所述语音数据进行情感分析，确定所述用户在发出所述语音数据时的音量和/或语速；
    根据所述音量和/或语速，获得所述用户的情感值；
    根据所述用户的情感值和预设的情感值与优先级映射关系，获得所述语音数据的优先级。
  4. 根据权利要求3所述的方法，其特征在于，所述根据所述音量和/或语速，获得所述用户的情感值，包括：
    对所述音量和/或语速进行数值处理，将数值处理的结果作为所述用户的情感值。
  5. 根据权利要求1-4任一项所述的方法，其特征在于，所述根据所述语音数据的优先级，将所述用户接入对应的服务通道，包括：
    将所述语音数据的优先级在预设的优先级与服务通道的映射关系中进行匹配；
    当匹配到人工通道时，将所述用户接入所述人工通道。
  6. 根据权利要求5所述的方法，其特征在于，还包括：
    根据所述语音数据，生成参考信息；
    将所述参考信息提供给所述人工通道对应的客服人员。
  7. 一种语音信息处理装置，其特征在于，包括：
    获取模块，用于获取用户的语音数据；
    分析模块，用于对所述语音数据进行情感分析，以确定所述语音数据的优先级；
    服务接入模块，用于根据所述语音数据的优先级，将所述用户接入对应的服务通道。
  8. 根据权利要求7所述的装置，其特征在于，所述获取模块具体用于：
    接收所述用户的通话请求，以与所述用户进行语音通话；
    向所述用户播放语音提示音；
    接收所述用户在所述语音提示音的提示下发出的所述语音数据。
  9. 根据权利要求7所述的装置，其特征在于，所述分析模块具体用于：
    对所述语音数据进行情感分析，确定所述用户在发出所述语音数据时的音量和/或语速；
    根据所述音量和/或语速，获得所述用户的情感值；
    根据所述用户的情感值和预设的情感值与优先级映射关系，获得所述语音数据的优先级。
  10. 根据权利要求9所述的装置，其特征在于，所述分析模块具体用于：
    对所述语音数据进行情感分析，确定所述用户在发出所述语音数据时的音量和/或语速；
    对所述音量和/或语速进行数值处理，将数值处理的结果作为所述用户的情感值；
    根据所述用户的情感值和预设的情感值与优先级映射关系，获得所述语音数据的优先级。
  11. 根据权利要求7-10任一项所述的装置，其特征在于，所述服务接入模 块具体用于：
    将所述语音数据的优先级在预设的优先级与服务通道的映射关系中进行匹配；
    当匹配到人工通道时，将所述用户接入所述人工通道。
  12. 根据权利要求11所述的装置，其特征在于，还包括：
    参考处理模块，用于根据所述语音数据，生成参考信息，并将所述参考信息提供给所述人工通道对应的客服人员。
PCT/CN2015/096241 2014-12-09 2015-12-03 语音信息处理方法及装置 Ceased WO2016091110A1 (zh)

Priority Applications (6)

Application Number Priority Date Filing Date Title
KR1020197008942A KR102284973B1 (ko) 2014-12-09 2015-12-03 음성 정보를 처리하기 위한 방법 및 장치
KR1020177018385A KR20170088997A (ko) 2014-12-09 2015-12-03 음성 정보를 처리하기 위한 방법 및 장치
EP15868151.0A EP3232651B1 (en) 2014-12-09 2015-12-03 Method and apparatus for processing voice information
US15/532,934 US10708423B2 (en) 2014-12-09 2015-12-03 Method and apparatus for processing voice information to determine emotion based on volume and pacing of the voice
SG11201704392QA SG11201704392QA (en) 2014-12-09 2015-12-03 Method and apparatus for processing voice information
JP2017530753A JP6538846B2 (ja) 2014-12-09 2015-12-03 音声情報を処理するための方法及び装置

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410748579.X 2014-12-09
CN201410748579.XA CN105744090A (zh) 2014-12-09 2014-12-09 语音信息处理方法及装置

Publications (1)

Publication Number Publication Date
WO2016091110A1 true WO2016091110A1 (zh) 2016-06-16

Family

ID=56106695

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/096241 Ceased WO2016091110A1 (zh) 2014-12-09 2015-12-03 语音信息处理方法及装置

Country Status (7)

Country Link
US (1) US10708423B2 (zh)
EP (1) EP3232651B1 (zh)
JP (1) JP6538846B2 (zh)
KR (2) KR20170088997A (zh)
CN (1) CN105744090A (zh)
SG (1) SG11201704392QA (zh)
WO (1) WO2016091110A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107277246A (zh) * 2017-06-16 2017-10-20 珠海格力电器股份有限公司 一种信息提醒方法及其装置、电子设备
CN107993674A (zh) * 2016-10-27 2018-05-04 中兴通讯股份有限公司 一种情绪控制方法及装置
CN110149450A (zh) * 2019-05-22 2019-08-20 欧冶云商股份有限公司 智能客服应答方法及系统

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106534598A (zh) * 2016-10-28 2017-03-22 广东亿迅科技有限公司 一种基于情感识别的呼叫平台排队系统及其实现方法
CN106791107A (zh) * 2016-12-22 2017-05-31 广东小天才科技有限公司 一种提醒方法及装置
CN107274893A (zh) * 2017-05-19 2017-10-20 四川新网银行股份有限公司 一种对客服系统智能调整在线排队优先级的方法
CN107393541B (zh) * 2017-08-29 2021-05-07 百度在线网络技术（北京）有限公司 信息验证方法和装置
JP7103769B2 (ja) * 2017-09-05 2022-07-20 京セラ株式会社 電子機器、携帯端末、コミュニケーションシステム、見守り方法、およびプログラム
CN107767869B (zh) * 2017-09-26 2021-03-12 百度在线网络技术（北京）有限公司 用于提供语音服务的方法和装置
CN108632475B (zh) * 2018-03-22 2021-08-17 平安科技（深圳）有限公司 语音留言方法、装置、计算机设备和存储介质
CN108876177A (zh) * 2018-06-28 2018-11-23 北京博数嘉科技有限公司 一种旅游咨询管理平台及管理方法
CN110970017B (zh) * 2018-09-27 2023-06-23 北京京东尚科信息技术有限公司 人机交互方法及系统、计算机系统
CN109618068B (zh) * 2018-11-08 2021-04-16 上海航动科技有限公司 一种基于人工智能的语音业务推送方法、装置与系统
CN109767240A (zh) * 2018-12-17 2019-05-17 中国平安财产保险股份有限公司 云客服分配方法、装置、计算机设备和存储介质
CN110289017A (zh) * 2019-06-06 2019-09-27 秒针信息技术有限公司 语音处理方法以及装置
CN112309586A (zh) * 2019-07-26 2021-02-02 深圳百诺明医说科技有限公司 提高医疗机器人推送信息与用户病症适配度的方法及系统
CN110890096A (zh) * 2019-10-12 2020-03-17 深圳供电局有限公司 一种基于语音分析的智能语音系统及方法
CN111161733B (zh) * 2019-12-31 2022-09-23 中国银行股份有限公司 一种智能语音服务的控制方法及装置
CN112040074B (zh) * 2020-08-24 2022-07-26 华院计算技术(上海)股份有限公司 基于语音声学信息的电话客服人员职业倦怠检测方法
WO2022054910A1 (ja) * 2020-09-14 2022-03-17 日本電気株式会社 接続装置、接続システム、接続方法等
CN112820323B (zh) * 2020-12-29 2023-06-16 平安银行股份有限公司 基于客户语音进行响应队列优先级调整方法及系统
CN115910110A (zh) * 2022-11-30 2023-04-04 杭州网筌科技有限公司 一种基于自然语言识别的政务服务系统
CN116580721B (zh) * 2023-07-13 2023-09-22 中国电信股份有限公司 表情动画的生成方法、装置和数字人平台

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2003073417A2 (en) * 2002-02-26 2003-09-04 Sap Aktiengesellschaft Intelligent personal assistants
CN101662549A (zh) * 2009-09-09 2010-03-03 中兴通讯股份有限公司 一种基于语音的客户评价系统及客户评价方法
CN101997996A (zh) * 2009-08-13 2011-03-30 华为技术有限公司 一种排队路由的方法及装置
CN102082879A (zh) * 2009-11-27 2011-06-01 华为技术有限公司 呼叫中心语音检测的方法、装置及系统
CN103413113A (zh) * 2013-01-15 2013-11-27 上海大学 一种服务机器人智能情感交互的方法

Family Cites Families (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5559927A (en) 1992-08-19 1996-09-24 Clynes; Manfred Computer system producing emotionally-expressive speech messages
US5860064A (en) * 1993-05-13 1999-01-12 Apple Computer, Inc. Method and apparatus for automatic generation of vocal emotion in a synthetic text-to-speech system
US6094635A (en) 1997-09-17 2000-07-25 Unisys Corporation System and method for speech enabled application
GB9723813D0 (en) * 1997-11-11 1998-01-07 Mitel Corp Call routing based on caller's mood
US6185534B1 (en) 1998-03-23 2001-02-06 Microsoft Corporation Modeling emotion and personality in a computer user interface
US6144938A (en) 1998-05-01 2000-11-07 Sun Microsystems, Inc. Voice user interface with personality
US6665644B1 (en) * 1999-08-10 2003-12-16 International Business Machines Corporation Conversational data mining
US7457397B1 (en) 1999-08-24 2008-11-25 Microstrategy, Inc. Voice page directory system in a voice page creation and delivery system
US7222075B2 (en) 1999-08-31 2007-05-22 Accenture Llp Detecting emotions using voice signal analysis
US6490564B1 (en) 1999-09-03 2002-12-03 Cisco Technology, Inc. Arrangement for defining and processing voice enabled web applications using extensible markup language documents
US6658388B1 (en) 1999-09-10 2003-12-02 International Business Machines Corporation Personality generator for conversational systems
US7140004B1 (en) 2000-06-13 2006-11-21 Tellme Networks, Inc. Method and apparatus for zero-footprint phone application development
US6757365B1 (en) 2000-10-16 2004-06-29 Tellme Networks, Inc. Instant messaging via telephone interfaces
US6795808B1 (en) 2000-10-30 2004-09-21 Koninklijke Philips Electronics N.V. User interface/entertainment device that simulates personal interaction and charges external database with relevant data
US20030007609A1 (en) 2001-07-03 2003-01-09 Yuen Michael S. Method and apparatus for development, deployment, and maintenance of a voice software application for distribution to one or more consumers
US7242752B2 (en) 2001-07-03 2007-07-10 Apptera, Inc. Behavioral adaptation engine for discerning behavioral characteristics of callers interacting with an VXML-compliant voice application
US7665024B1 (en) * 2002-07-22 2010-02-16 Verizon Services Corp. Methods and apparatus for controlling a user interface based on the emotional state of a user
US6959080B2 (en) * 2002-09-27 2005-10-25 Rockwell Electronic Commerce Technologies, Llc Method selecting actions or phases for an agent by analyzing conversation content and emotional inflection
US7676034B1 (en) 2003-03-07 2010-03-09 Wai Wu Method and system for matching entities in an auction
JP4197344B2 (ja) * 2006-02-20 2008-12-17 インターナショナル・ビジネス・マシーンズ・コーポレーション 音声対話システム
US20080086690A1 (en) 2006-09-21 2008-04-10 Ashish Verma Method and System for Hybrid Call Handling
JP2009182433A (ja) 2008-01-29 2009-08-13 Seiko Epson Corp コールセンターの情報提供システム、情報提供装置、情報提供方法及び情報提供プログラム
CN101741953A (zh) * 2009-12-21 2010-06-16 中兴通讯股份有限公司 一种通话时应用卡通动画显示语音信息的方法和装置
CN102142253B (zh) * 2010-01-29 2013-05-29 富士通株式会社 语音情感识别设备及方法
US20130051545A1 (en) * 2011-08-25 2013-02-28 Bank Of America Corporation Call center system for dynamic determination of appropriate representative
CN102623009B (zh) * 2012-03-02 2013-11-20 安徽科大讯飞信息科技股份有限公司 一种基于短时分析的异常情绪自动检测和提取方法和系统
CN102802114B (zh) * 2012-06-20 2015-02-18 北京语言大学 利用语音进行座席筛选的方法及系统
CN103811009A (zh) * 2014-03-13 2014-05-21 华东理工大学 一种基于语音分析的智能电话客服系统
US10101974B2 (en) * 2014-07-31 2018-10-16 Angel.Com Incorporated Contact center application creating using reusable program modules

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2003073417A2 (en) * 2002-02-26 2003-09-04 Sap Aktiengesellschaft Intelligent personal assistants
CN101997996A (zh) * 2009-08-13 2011-03-30 华为技术有限公司 一种排队路由的方法及装置
CN101662549A (zh) * 2009-09-09 2010-03-03 中兴通讯股份有限公司 一种基于语音的客户评价系统及客户评价方法
CN102082879A (zh) * 2009-11-27 2011-06-01 华为技术有限公司 呼叫中心语音检测的方法、装置及系统
CN103413113A (zh) * 2013-01-15 2013-11-27 上海大学 一种服务机器人智能情感交互的方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3232651A4 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107993674A (zh) * 2016-10-27 2018-05-04 中兴通讯股份有限公司 一种情绪控制方法及装置
CN107277246A (zh) * 2017-06-16 2017-10-20 珠海格力电器股份有限公司 一种信息提醒方法及其装置、电子设备
CN107277246B (zh) * 2017-06-16 2018-12-28 珠海格力电器股份有限公司 一种信息提醒方法及其装置、电子设备
CN110149450A (zh) * 2019-05-22 2019-08-20 欧冶云商股份有限公司 智能客服应答方法及系统

Also Published As

Publication number Publication date
CN105744090A (zh) 2016-07-06
US10708423B2 (en) 2020-07-07
JP2018501721A (ja) 2018-01-18
EP3232651A1 (en) 2017-10-18
JP6538846B2 (ja) 2019-07-03
KR102284973B1 (ko) 2021-08-03
KR20190037363A (ko) 2019-04-05
US20170346947A1 (en) 2017-11-30
EP3232651B1 (en) 2025-06-04
EP3232651A4 (en) 2018-05-09
SG11201704392QA (en) 2017-06-29
KR20170088997A (ko) 2017-08-02

Similar Documents

Publication Publication Date Title
WO2016091110A1 (zh) 语音信息处理方法及装置
US10057419B2 (en) Intelligent call screening
CN110661927B (zh) 语音交互方法、装置、计算机设备及存储介质
CN107623614B (zh) 用于推送信息的方法和装置
US10134395B2 (en) In-call virtual assistants
JP7682950B2 (ja) 保留を管理するための方法および装置
CN107818798A (zh) 客服服务质量评价方法、装置、设备及存储介质
US20060122840A1 (en) Tailoring communication from interactive speech enabled and multimodal services
CN108777751A (zh) 一种呼叫中心系统及其语音交互方法、装置和设备
EP3158719A1 (fr) Procede et systeme de filtrage d'appels telephoniques entrants indesirables
WO2016045479A1 (zh) 客服电话处理方法及装置
US20130124631A1 (en) Apparatus, system, and method for digital communications driven by behavior profiles of participants
CN113067952B (zh) 用于多台机器人的人机协同无感控制方法和装置
US20120101873A1 (en) Method and apparatus for dynamic communication-based agent skill assessment
JP2021196462A (ja) 電話応対装置
US11138379B2 (en) Determination of transcription accuracy
CN113327582B (zh) 语音交互方法、装置、电子设备及存储介质
US10972606B1 (en) Testing configuration for assessing user-agent communication
US8085907B2 (en) Dynamic control of voice mail delivery sequence
CN120375821A (zh) 基于语音网关的智能语音交互方法及装置
CN116684392A (zh) 电脑和手机的通讯方法、装置、电子设备及存储介质
CN117854521A (zh) 声音信息保护方法及装置、存储介质、电子装置
CN114598773A (zh) 一种智能应答系统及方法
JP2014164077A (ja) 音声処理サーバ

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: 15868151

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 11201704392Q

Country of ref document: SG

WWE Wipo information: entry into national phase

Ref document number: 15532934

Country of ref document: US

REEP Request for entry into the european phase

Ref document number: 2015868151

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2017530753

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 20177018385

Country of ref document: KR

Kind code of ref document: A

WWG Wipo information: grant in national office

Ref document number: 2015868151

Country of ref document: EP