A METHOD AND SYSTEM FOR SUPPORT WITH VOICE INTERFACE
Field of the invention
The present invention relates to a method and a system for computerized support.
Background of the invention
In all times, the most natural and obvious way for humans to communicate is to use sounds, i.e. to use a voice interface. In many cases, this is also preferred, since it makes it possible to communicate while doing other things with the hand, looking at other things, etc.
However, it has proven very difficult to use such voice interfaces in the communication between humans and machines, such as different types of computerized systems. Accordingly, in spite of the inherent advantages of voice interfaces in many situations, visual interfaces are used instead.
Especially in situations where a number of questions need to be responded by the user, a voice interface would be advantageous, and some attempts have also been made in using voice interfaces in such systems.
For example, WO 88/05918 discloses an apparatus for aiding technicians in diagnosing malfunctioning devices.
The apparatus comprises a database contains diagnostic data including the possible causes of the faulted status of the device, corresponding operation conditions, and information about corresponding suitable repair actions.
Diagnosis is performed by matching the input, e.g. provided by a voice interface, with the possible fault conditions stored in the database.
Further, US 5 660 176 discloses a system for computerized medical diagnosis. A patient responds to a number of questions, and based on the responses a
diagnosis is made. A database may also be used for storage of the patients medical record.
Still further, EP 0 342 859 discloses a patient assistance system, which provides interactive medical diagnosis and treatment. Some of the communication between the system and the patient is directed through a voice interface.
However, such known systems using voice interfaces are suffering from several problems. For example, the information input is tedious and cumbersome, due to the large amount of questions needed to be responded. In order to be able to handle different situations, at least some questions are inevitably irrelevant for each situation, which could be distracting and annoying for the user. Still further, the systems are normally relatively inflexible, making it difficult to amend and alter the system, for adaptation to new support situations, new needs, etc.
Summary of the invention
It is therefore an object of the present invention to provide an improved method and system for computerized support using voice interface.
This object is achieved with the method and system as defined in the appended claims.
The invention relates to a method for computerized support comprising the steps of: providing at least one input device, at least one support server, a database connected to said server, and a communication network connecting said input device and said support server, said input device providing a voice interface between the input device and a user; and in the server: receiving information about an object for which support is requested from the input device; determining, based on said information, the type of
object for which support is requested; demanding and receiving a response for at least one question from the user through the voice interface; demanding and receiving a response for at least one further question through the voice interface, said further question being based on the response of the previous question, and repeating this step at least once; determining, based on said responses and pre-stored information in said database, a probable fault condition for the object for which support is requested; and determining, based on pre-stored information in said database associated with said determined probable fault condition, an action to be taken for alleviation of said fault condition. This support method could be used for a wide variety of different products and objects, and is especially suitable for technical devices, and especially electrical devices, such as computers, telephones, TV-sets, video equipment, audio equipment, copiers, etc. However, it may also be used for other types of technical equipment, such as cars, bicycles, vacuum cleaners, etc, or even for nontechnical objects, such as forms to be filled in or the like. The support service may be easily accessed by e.g. dialing a telephone number on an ordinary phone, and the telephone number may be clearly indicated on the object, preferably together with the identity code for the object. Hereby, efficient support service may be provided for almost every type of products around the clock every day in the week. The support system is further relatively inexpensive and easy to set up, and could easily be changed or improved in different ways over time. The use of questions which depends on the earlier responses makes the process of determining the problem very efficient and increases the reliability. The reception of information about the object for which protection is sought makes it possible both to use the same support system for many different types of objects, and to personalize the
support service to a large extent, making it more efficient and reliable. Preferably, the information about the object comprises an identification code for said object, and most preferably the identification code is a unique code associated with said object for identification.
The determined action could be performed by an operator of said object for which support is requested, and the method could then further comprise the subsequent step of demanding and receiving a response for at least one further question related to the result of the performed action. Hereby, the user could easily and effectively be guided through the remedy or repair process . It is also possible that the object for which support is required is connected to the communication network, wherein the method could comprise the further step of executing the determined action automatically through said communication network. This renders the process even more effective.
At least some, and preferably all, of said questions are further formulated to narrow the scope of possible fault conditions which are pre-stored in said database. Hereby, the transformation between speech and data understandable by the support server becomes more easy. For the same reason, at least some, and preferably all, the questions are further formulated to provide a limited set of possible responses.
In an embodiment, the method further comprises the. steps: receiving information about the user requesting support ; using said received user information for user personalization of at least one further step. Hereby, the support service could be personalized for each user or group of user, which makes it more efficient and reliable. At least one of the following steps could
preferably be personalized: the determination of the type of object for which support is requested; the choice of questions to be responded; the determination of a probable fault condition; and the determination of an action to be taken for alleviation of said fault condition.
Preferably, the received user information is used to access additional user information associated with said user from a user information database connected to said server.
The received information about the user requesting support could additionally or alternatively be used for controlling payment from the user for the support service.
The invention further relates to a corresponding system for computerized support comprising at least one input device, at least one support server, a database connected to said server, and a communication network connecting said input device and said support server, said input device providing a voice interface between the input device and a user, said support server comprising: means for identifying the type of object for which support is requested from information about said object provided by the input device; means for generation of questions based on said identified type of object and information in said database, and, except for the first question, on a response of the previous question, wherein said questions and responses are to be transmitted through the voice inter ace; means for determining, based on said responses and pre-stored information in said database, a probable fault condition for the object for which support is requested; and means for determining, based on pre-stored
information in said database associated with said determined probable fault condition, an action to be taken for alleviation of said fault condition.
Further scope of the applicability of the present invention will become apparent from the detailed description given hereinafter. However, it should be understood that the detailed description and specific examples, while indicating preferred embodiments of the invention, are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description.
Brief description of the drawings
For exemplifying purposes, the invention will be described in closer detail in the following with reference to embodiments thereof illustrated in the attached drawings, wherein: Fig 1 is a schematic illustration of a support system according to an embodiment of the present invention; and
Fig 2 is a flow chart illustrating a support method according to an embodiment of the present invention.
Description of preferred embodiments
A system for computerized support according to an embodiment of the invention is illustrated in fig 1. The support system comprises an input device 1, and preferably several input devices, such as telephones, and preferably cellular or mobile phones, communicating through a wireless network 6. However, telephones connected to the ordinary telephone network (PSTN) could also be used. Further, it is also possible to use other types of input devices. The input device comprises a voice interface for receiving and transmitting speech information. To this end, the input device preferably
comprises a microphone for speech reception from the user and a speaker, earphone or the like for sound transmission to the user.
The system further comprises at least one support server 3. The server could for example comprise a conventional network server. The server is connected to at least one, and preferably several databases 4. The databases could either be integrated with the server, or be arranged outside the server but connected to the same. Further, in case of several databases, they could either be stored in separate storage units, or in different memory areas on the same memory or storage unit .
The input device 1 and the server 3 are connected through a communication network. The communication network could comprise a wire less communication link, a permanent link or combinations of the two. Preferably, the support server is connected to a wide area network 5, such as the Internet . The input device could be directly connected to the same network 5, but preferably, it is indirectly connected to the network through an intermediate unit 2. The connection 6 between the intermediate unit 2 could be an ordinary telephone connection, either through the permanently switched telephone network (PSTN) , or through a wireless network, such as the GSM. For transformation between data and speech, a speech synthesizer and a speech recognition unit are preferably used. These units could be arranged in the input unit or in the support server, but preferably they are both arranged in the intermediate unit, in case a such is present.
The support server comprises means for identifying the type of object for which support is requested from information about said object provided by the input device. The information about the object could preferably comprise an identification code for said object, and preferably a unique code associated with said object for
identification. However, it is also possible to have non- unique codes, being general for certain types or categories of objects, or groups of such objects. The codes could be entered as data, through a keypad, keyboard or the like on the input device, or through the voice interface. However, alternatively the information about the object or type of object could be entered in other ways, such as a general description in plain words, or by responding to questions about the object provided by the support server. The object information could also be communicated automatically by the input device when connecting to the network and the support server.
Based on the information about the object, the type of object is determined, and perhaps event the specific object in question. This determination is based on the input data in combination with pre-stored data in the database. In case of a code input, the database could comprise an association, such as a list or a mapping, correlating the codes with pre-stored information about the corresponding objects.
The support server further comprises means for generation of questions based on said identified type of object and information in said database, and, except for the first question, on a response of the previous question. The first question is preferably based on the information about the object, and the set of possible faults and problems related to this object stored in the database .
Several, and preferably all, the questions are preferably formulated to narrow the scope of possible fault conditions which are pre-stored in said database, wherein the number of possible faults or problems diminishes with every response provided by the user. By means of said responses and the information stored in the database, the most probable fault condition could then be automatically determined.
The questions are further preferably formulated to provide a limited set of possible responses, such as "yes" or "no", a choice among a defined set of alternatives, a numerical value, or the like. Hereby, the speech recognition becomes easier and more reliable, and the automatic drawing of conclusions based on the response is greatly facilitated.
Further, the voice interface preferably uses a predetermined limited speech grammar, which also facilitates the speech recognition and the computerized handling based on the responses. For example, voice interfaces such as the per se previously known Voice XML could be used. A person versed in the art realizes that many other types of speech grammars could be defined and used to achieve the purpose of the present invention. The support server also comprises means for determining an action to be taken for alleviation of the determined probable fault condition. This could preferably be accomplished by associating one or several such actions to each problem or fault defined in the database. The determination could then be made by mapping or listing the relation between the problems or faults and the proposed actions.
In case the object for which support is sought is connected to the network and accessible by the support server, the action could executed on the object automatically. For example, software related problems may be solved remotely and handled automatically in this manner. The action may also comprise one or several steps to be performed manually by e.g. the user or any other operator of the object for which support is sought. Instructions for such actions may also be communicated through the voice interface . In case steps are to be executed manually, check-up questions are preferably formulated by the support server, in order to receive a response related to the result of the performed action.
Hereby, the operator could easily be guided through the remedy process, under almost complete control of the support server. The check-up questions also provides an opportunity to correct difficulties and handling errors, and, if possible, choose alternative solutions when needed.
The support server may also comprise means for identifying information about the user, and preferably the identity of the user. The information about the user could preferably comprise an identification code for the user, and preferably a unique code associated with the client. However, it is also possible to have non-unique codes, being general for different groups of users, such as users of different skill levels, users associated with different companies, users from different countries and the like. The codes could be entered as data, through a keypad, keyboard or the like on the input device, or through the voice interface. However, alternatively the information about the client or user could be entered in other ways, such as a general description in plain words, or by responding to questions about provided by the support server. The user information could also be communicated automatically by the input device when connecting to the network and the support server, by means of so called cookies or the like.
The user information may be used for personalization and adaptation of the process of determining the object or type of object for which support is requested, the diagnostic process of determining the problem or fault of the object, such as the formulation of questions and drawing of conclusions based on the responses, and/or the process of repair and remedy of the identified fault. For example, the user information may be used to access pre- stored data about the user or group of users from a user information database. This pre-stored data could comprise information about previous support events, equipment in
the possession of the user, competence of the user, etc. Hereby, the support may be personalized to make it more efficient. By continuously updating the database, the support system could easily self-improving, whereby it will become more efficient over time.
Additionally, or alternatively, the user information may be used for accounting purposes. Hereby, it becomes possible for the support provider to receive payment for the support service provided to the user. The user information may in this case be used for automatic or semi-automatic debiting for the provided service.
It is also possible to provide payment for the support service by charging the communication, such as communication on the telephone network. The means and functionality of the support server discussed above may be software implemented, but may also be implemented in hardware.
A method according to a preferred embodiment is illustrated in fig 2.
In a first step SI the user connect to the support server through the input device. This could e.g. be done by dialing a certain number. Thereafter, it is determined whether the ID for the object is known, step S2. If the ID is known, it is provided to the support server, step S3, or else, the identity is determined by use of questions formulated by the support server, step S4.
Then, it is determined whether the identified support object could be handled by the support server, step S5. In case the support situation could not be handled by the computerized system the user is process is ended and the user is disconnected or transferred to another support system, step S6. Otherwise, questions is formulated and asked to the user in order to identify and determine the problem or fault of the object as is discussed in more detail above, step S7.
Thereafter, it is concluded whether the problem could be solved on spot, or if qualified assistance is needed, step S8. If so, the user is informed about the situation and is instructed to seek qualified assistance, step S9. If the problem is manageable, it is then determined whether the problem could be solved automatically, step S10. If so, the remedy actions are executed automatically, step Sll. If manual steps are required, an instruction for those steps are generated and transferred to the user, as is discussed in more detail above, step S12. It is then confirmed that the problem is solved, for example by check-up questions, step S13. If the problem remains, or if there are any other problems, the process is returned to either step S6 or Sll, depending on the situation. When the problems are solved, the process is ended in step S14.
It should be appreciated, that several of the above discussed steps are optional, or could be performed in a different order. It is further possible to add additional steps in the process.
The support system and method of the invention is especially suitable for technical devices, and especially electrical devices, such as computers, telephones, TV- sets, video equipment, audio equipment, copiers, etc.
However, it may also be used for other types of technical equipment, such as cars, bicycles, vacuum cleaners, etc, or even for non-technical objects, such as forms to be filled in or the like. The support service may be accessed by dialing a telephone number on an ordinary phone, and the telephone number may be clearly indicated on the object, preferably together with the identity code for the object. Hereby, efficient support service may be provided for almost every type of products around the clock every day in the week. The support system is further relatively inexpensive and easy to set up, and
could easily be changed or improved in different ways over time.
The invention has now been described by means of embodiments. However, many alternative embodiments are possible. For example, different types of input devices may be used, the support server may be implemented in different ways, the support system may be used for many different types of products and objects, etc. Such and other closely related alternatives must be considered to be within the scope of the present invention as it is defined in the appended claims.