NZ530400A - An intelligent messaging server for bridging text and non-text services - Google Patents

Abstract

An intelligent messaging server for bridging text and non-text capable devices is disclosed. The intelligent messaging server includes (a) a text message gateway that is configured to send and receive text messages to a terminal, (b) a media interface configured to send and receive real time data to a real time communications network, (c) a session controller configured to control the establishment of a real-time media communications session between a real time media communications interface and a media terminal, (d) a text to speech processor, (e) a message store, (f) a media store, (g) a media streamer configured to stream media data in real time from a media data source to a real time media interface, (h) a media analyser configured to determine the class of media terminal to which the system is connected, (i) a signal analyser configured to detect and interpret responses supplied by the user at a media terminal, (j) an activity log to hold a record of activities and events during the processing of text messages, and (k) an automated delivery manager.

Description

530400 COMPLETE SPECIFICATION TITLE AN INTELLIGENT MESSAGING SERVER FOR BRIDGING TEXT AND NON-TEXT SERVICES- ABSTRACT A system for delivery of bi-directional messages between text-capable devices and text-incapable devices is disclosed. A text-capable device is used to send a text message to a destination that cannot communicate using text. The system detects the type of destination and translates the text into a form that can be understood by the device and, if necessary, by the attendant at that device. Abbreviations commonly used in text communications are expanded. The data may be modified, for example by the use of different voices, according to the wishes of the sender or receiver. The system may, at the receivers request, deliver the message in different ways in order that the message can be understood; for example, the message may be read more slowly, or words may be spelt out. The sender may identify that the message requires a response and may also indicate the type of response required, for example, an enumeration, yes/no, a. number, a date or a time. The receiver may respond to the message and the response is translated back to the text form. The receiver may elect to respond by returning the call directly without terminating the delivery call, wherein the sender or receiver may pay for that call. If the user is not located then the system may escalate the contact by calling again or trying other addresses. The time of delivery may be specified by the sender and further constrained by the receiver to avoid antisocial times. The target user may be IINTELLECTUAL PROPFRTY OFFICE OF M.Z - 5 APR 2005 RECEIVED authenticated by a number of means, such as PIN or biometric methods. The receiver may elect only to receive from nominated senders or screen senders at the time of delivery. A receiver may blacklist a sender during the process of receiving a message from that sender. Advertising or branding may be used as a means of subsidising the cost of the call and in such cases, the receiver may be offered the opportunity to be connected directly to an automated or live agent for the purpose of purchasing or receiving additional information about a product.

FIELD

[0001] The invention relates to messaging services. More particularly, the invention relates to techniques for enabling rich bi-directional communications between text-capable and non-text-capable devices. The invention also addresses the problem of confirming that the message has been received and understood.

BACKGROUND

[0002] Throughout this disclosure, the term "advertising" is used in its most general sense and is intended to mean delivery of information where there is benefit to either or both parties and where that benefit may be commercial or otherwise.

[0003] The object of the present invention is that of ensuring that a message between text-capable devices and voice-only devices are delivered and acknowledged. Unified communications systems perform many of these functions, however they fail to address several problems.

[0004] Some communications can be achieved more effectively by a non-real-time request and response over media such as SMS, Instant Messaging or email. Examples of cases where such non-real-time delivery is advantageous include time zones and scheduled absence, the reluctance of a sender to disturb the receiver, or simply that it is an inconvenient time to engage in the human protocols of 2-way communication.

[0005] Mobile telephones are the most ubiquitous text-capable devices. However, because they also support voice calling, the user is constantly faced with an option to send a text message or make a voice call.

[0006] There are many reasons a user will choose text over voice. Some reasons are related to the inherent mobility of cellular telephones. For example, the need for privacy and discretion in the presence of other persons, or the unreliability of delivery and the likelihood of the call being redirected to voice messaging system. However, it is the issue of cost is the primary driver behind the preference for text messaging over voice for within the younger community.

[0007] In other cases, this is the text messaging is the only method available. For example, when the sender does not have access to a voice capable device, the target telephone does not answer, is busy or is connected to a FAX device.

[0008] Social and cultural factors must also be considered. Text communications is fashionable with the young. Parents however, are not as adept or have less inclination to learn text messaging and may not use cellular telephones at home. It would be of considerable advantage to Parents if they could increase communications with young people by using means that are most acceptable to each party.

[0009] Given that text messaging is often a preferred, and sometimes the only, way to communicate, it follows that it would be an advantage if the range of possible participants could be extended to non-text capable devices.

[0010] Delivering text to a non-text capable device requires translation of the data into a form supported by that device. In the case of Text sent to a telephone this requires Text To Speech (TTS) technology. However, where the message must be entered on a restrictive interface or the message length is limited, abbreviations are commonplace.

[0011] It is not sufficient for a system to create a list of known abbreviations or even to leam these abbreviations since they are often highly specific to a given context. Worse still, they may only be only meaningful to the parties concerned. For example, " the text BFN" might represent the initials of a colleague, the name of a local restaurant or 'Bring food now'.

[0012] Intelligent translation of abbreviated text into speech can never be resolved fully by automated processes. It would be an advantage if the delivery process were able to offer different options to help the receiver understand the message. Strategies include speaking more slowly, increasing the volume or speaking the text, or parts of it, word-by-word or letter-by-letter. Reading the message letter-by-letter is slow and cumbersome, but it results in the message being delivered to the same level of accuracy that it was entered. If it is still unintelligible then it will not be due to the system but to the composer's lack of clarity. By playing the message and offering progressively more options, the user is in ultimate control and can determine the level of accuracy they are prepared to accept.

[0013] A further problem with many systems for message delivery is that the target address may be that of a FAX machine. It would be an advantage if in these cases the message could be converted into a fax message, where the message may include other opportunistic messaging such as advertising or promotions. It would be a further advantage if the system notified the sender that a fax wets left or even used rules provisioned by the sender, receiver or both to determine whether other destinations should be tried. [0014j Answering machines provide another problem for automated messaging systems. It would be an advantage if, upon encountering an answering machine, a message could be left on that answering machine. It would be a further advantage if the system notified the sender that a recorded message was left.

[0015] Even in the event that a human operator is contacted, there are several issues to be resolved. A common problem with all landline telephone devices is that often, the caller may wish to speak with a specific person at that address but another person answers the call. It would be an advantage if the system were able to determine if the user was present using a number of means. The means could include a simple affirmation, a PIN or a Biometric verification process such as speaker verification.

[0016] It would also be an advantage if, at the callers request, the call is not completed until the required person is present, or the system could pass a short message, or request, to be delivered person required.

[0017] When a user other than the intended target receives the call, it would be an advantage if a different message could be left, for example, one that indicated that the message could be retrieved. Alternatively a dialog could be executed in order to ascertain the best time for another attempt at delivery and/or an alternate address to be used.

[0018] It may be that the receiver is unable to take details of the call and would prefer to be called at a later time. It would be an advantage if the system could call the user again at a later time or at a time that the user requests.

[0019] It would be a further advantage if, in the case that the message delivery failed, specific actions could be specified by the receiver, the sender or both, where the actions might include calling alternate numbers or retrying the same number at an interval which is selected according to the likelihood of the target user being present at that time. For example, if the called party is busy, it is likely that a call a few minutes later might succeed whereas a call that went unanswered is less likely to succeed if re-scheduled for a short period of time later.

[0020] A disadvantage of messaging services is that it is typically unidirectional. It would be of considerable advantage if the sender could solicit a response to a question, where the response could be one of a number of different types that are selected by the sender. The response types would include cardinal numbers, ordinal numbers, enumerations, yes/no, time, money etc. It would be an advantage if the response could be entered by Speech recognition and other means in addition to DTMF.

[0021] An advantage of a message over a real time call is that the communication transaction does not require the simultaneous presence of the two parties. This is particularly useful when time zones are different, the time that the message is sent is antisocial to the receiver or the sender is aware that the receiver is not available at the time of sending. A system that allows the sender to select the time that delivery should happen will be of significant advantage.

[0022] The receiver of messages is may wish not to receive telephone messages at any time of day. It would be an advantage if the receiver could specify when messages could and could not be delivered. It would be a further advantage if urgent messages could be identified and the receiver were able to provision the system to selectively handle urgent and non-urgent messages.

[0023] The receiver may be the billed party for such communications. That is, the receiver pays all or part of the cost of delivery of the message; the payment being on a transaction or subscription bases. It would therefore be an advantage if the receiver could selectively allow messages to be received by a landline number by provisioning those that are allowed to send messages to that number or by allowing the receiver to selectively, before hearing the message, choose to reject or accept messages. In the case that a message is rejected, it would be a further advantage if the sender could be blacklisted by the receiver at the time of the call.

[0024] The ability to send messages to abbreviated numbers would be an advantage. In the case of a closed user-group, where the members are nominated by a receiving landline, it would be an advantaged if the system could be provisioned to map one abbreviated number to a home phone number. It would be a further advantage if many different groups could share the abbreviated number and if each member of the group could use a different abbreviated number.

[0025] When receiving messages from telephone users, the calling telephone number often used to announce the identity of the sender. This is very difficult, particularly where numbers are long and similar, as with many cellular telephone numbers. It would be an advantage if, in the case that the receiver is provisioned, the system could detect particular senders and announce them by recorded or synthesised voice. It would further add to the experience if different synthesized voices, or voice parameters could be assigned to each sender to a given receiver, in order to provide additional emphasis.

Even a simple male/female setting would still add significant utility.

[0026] In cases where the receiver does not have the opportunity to select voices, it would be an advantage if the sender could make choices as to the voice, even if it were only between a male and female voice.

[0027] Although the technology is far from ubiquitous at the present time, it will be possible that a user can use their own voice as a model from which to generate synthesized speech. This will be a considerable advantage to the receiver.

[0028] The entertainment value or the message content itself can be augmented by the addition of audio that provides a backdrop or a mood. It would be an advantage if the audio could be mixed with the message as it is delivered, in order to provide additional entertainment or semantic value to the message, wherein the content that is mixed could be chosen by the sender or receiver.

[0029] On many occasions, text communications will become too limited for the communication transaction and a live telephone call would be the best medium. It would be of considerable value if the receiver could elect to be connected to the sender without abandoning the call or having to type in the number of the sender.

[0030] In the case that the receiver calls the sender back, it would be an advantage if the billing of that call could be flexible. That is: a. the sender may be asked by the receiver to accept the charge, b. the sender may offer to receive the charge in the initial message, c. the sender may explicitly request a call and offer to pay the charge in the initial message, d. the receiver may have to, or choose to, accept the charge.

[0031] Any service can be funded by advertising and branding in order to subsidise or eliminate costs to the users. Advertising may also be used to the benefit of the service provider as a means of creating opportunities to increase revenue. It would be of advantage to the users and providers of such a service if promotional messages could be played to users. It would also be useful if the promotional messages could be tailored to the receiving user.

[0032] Any advertising medium is only attractive to the degree that it is able to convert to sales, and that can demonstrate that is does so. It would be an advantage if the dialog was designed to intrude as little as possible but allow the user to elect to get more information, and that the statistical data be made available to the advertiser.

[0033] Converting an advertising message into a sale requires contact with the purchaser. It would be a considerable advantage if the dialog could result in the caller being connected to a live or automated agent that is able to complete the sale.

[0034] In some cases a sender may wish to send the message to more than one message at a time. It would be an advantage if a message could be sent to more than one destination without need for reentering the message.

[0035] It would be an advantage if the purpose of the message itself could be advertising or promotion.

[0036] The invention described overcomes all these problems and provides a platform for rich communications between users of text-capable and non-text devices.

BRIEF DESCRIPTION OF THE DRAWINGS j0037] The invention is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings in which like reference numerals refer to similar elements.

[0038] Figure 1 is one embodiment of a Text gateway.

[0039] The Text gateway (120) comprises a number of storage areas, a Message Delivery System (129) and communicates with several external entities.

[0040] The external communications are with: a. System Configuration Devices (150), which primarily control the system and populate or view the data stores, b. External Data Feeds (130) which are used as media to be mixed with the data delivered, c. Contact Centres (140), used to handle enquiries about advertised products or provide assistance, d. Real Time Communications Networks (110), through which the system communicates with audio terminals e. One or more Text Message Gateways (160) that allow the system to receive text messages sent to non-text-capable devices and enable responses to be sent to text-capable devices.

[0041] Figure 2 is one embodiment of a message delivery system

[0042] The Message Delivery System (129) is comprised of software and hardware devices. The Message Delivery System (129) receive and stores text messages, contacts the target user, detects the type of device and controls a number of text expansion and synthesis processes, media processing resources and the flow of data between resources, in order to deliver the message and solicit responses from the target user. [0043| Treble lines denote real-time data streams, single lines denote non-real-time data transfer and fine dotted lines denote control paths.

[0044] The figure omits the devices and processing components required to perform translation of text to FAX and to send a FAX. The extensions required will be self-evident to one with experience in the art.

[0045] Figure 3 shows the basic steps for delivering text messages non-text devices

[0046] The Message Delivery Manager (210) is responsible for coordinating the media and call processing resources in order to deliver the message to a non-text device and transmit a response to the initiator.

[0047] The diagram does not show the additional steps that are involved in processing the various special options that the sender or receiver may have requested, including the option of connecting the two parties in a real time session. Nor does it show the additional steps of delivering an advertising or promotional message, or transferring the user to an ACD (410) for assistance or to find out more about an advertised product.

DETAILED DESCRIPTION

[0048] Techniques for improving the ability for users of text-capable to communicate with users of non-text devices are disclosed.

[0049] In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the invention. It will be apparent, however, to one skilled in the art that the invention can be practiced without these specific details. In other instances, structures and devices are shown in block diagram form in order to avoid obscuring the

[0050] Reference in the specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment.

[0051] The techniques described herein are applicable to any real time communications medium or class of terminating device. For example, the communications medium may be 2-way radio, an instant message session, Push-to-talk session or a conventional telephone call.

[0052] The techniques described herein are applicable to any text capable terminating device, for example, the terminal may be an instant message client, email client, TTD device. Push-to-talk session or a conventional telephone call.

[0053] The techniques described herein may serve as part of a solution for 'deaf-relay' systems.

[0054] The techniques described herein are also applicable to any method of recording, storing, transmitting, transforming and rendering of data. For example, olfactory in formation could be, delivered to a capable device or, alternatively, rendered as text captions and displayed on a video screen at the receiving terminal. ]0055| In the component diagrams above, the communications lines between components do not signify a single transport or any limitation on the partitioning of components. In one embodiment, the communications paths are across network protocols and the processing entities are stored program computer processors. In another embodiment, the communications paths are internal inter-process communications (IPC) and the processing elements are processes sharing a single or multiple stored program computer processors.

[0056] With reference to Figure 1, the Text Gateway (120) interfaces with System Configuration Devices (150) for the purposes of administering the system, entering data into storage locations and viewing the contents of storage locations. System Configuration Devices (150) include, but need not be limited to, Web browsers, telephone IVR applications or automated entry devices using a protocol that communicates with the Message Delivery System (129).

[0057] External Data Feeds (130) provide digital data in a form that can be sent to a real time stream, however, the stream need not arrive in a manner that is regarded as real time, that is, the delay variation need not be acceptable for direct transmission without buffering in order to ensure no underflow condition occurs. External Data Feeds (130) may be mixed, by summation, with the main message delivery dialog to provide a backdrop or mood for entertainment or in order to enhance the experience.

[0058] One or more Contact Centers (140) may be used in order to allow the target user to be transferred to a live agent. The Text Gateway (120) communicates with the ACD (141) to transfer information known about the target user for display on an Agent Data Terminal (413) and, at the same time, the Text Gateway (120) transfers the call to the Agent Media Terminal (142). The transfer may be achieved by making a second call and bridging them, or by using a transfer feature made available by the Real Time Communications Network (110).

[0059] The Real Time Communications Network (110) provides a media transport for terminals to communicate. The term "real time" is relative to the type of media being transmitted. The requirements of different media vary widely. Voice communications require latencies of less than 150 milliseconds to prevent disturbance to the natural cadence of human conversation. Instant Messaging clients and FAX devices can handle longer delays.

[0060] The Real Time Communications Network (110) has 2 interfaces. A session control interface (single line) and a data stream interface (treble line). The control interface may be used to initiate, terminate and manage calls (sessions). Some Real Time Communications Networks (110) are able to transfer the media stream to another terminal directly.

[0061] The Text Message Gateway (160) appears to the Real Time Communications Network (110) as a terminal.

[0062] The Text Message Gateway (160) provides an interface to the text-capable devices. In some cases the text capable devices are also real time devices (such as cellular telephones). In other cases, the Text Message Gateway (160) receives messages from Text Terminals (163) via a Data Network (161).

[0063] The Text Gateway (120) holds data in a number of storage locations. Billing information is placed into the Billing Information Store (125). This information includes a record for all activity including, but not limited to: a. date, b. time, c. addresses of parties to the call if any, d. activity type, e. account identification, f. duration of activities,

[0064] The Billing Information Store (125) may be viewed by System Configuration Terminals (150). It will also be used by an external billing system (not shown).

[0065] In one embodiment, the Billing Information Store (125) is not required and therefore is omitted.

[0066] The User Group Store (123) contains the identities of the members of the each group as well as information about the group itself. An example of a User Group is a Land-Line Terminal (112) user that allows a number of Wireless Terminals (111) to send Text messages at no charge and, using techniques outlines in New Zealand Provisional Patent Application 527737, to allow the sharing of'free to text numbers'. In this way, families, companies or other groups can allow text as a method of communicating with a single location at no cost to the sender.

[0067] Data is entered into, and retrieved from the User Group Store (123) using System Configuration Terminals (150). In addition, the User Group Store (123) is accessed and modified by the Message Delivery System (129).

[0068] The User Preference Store (122) holds preferences of the sender or receiver. These are used to reduce the complexity of the communications process by storing information that would otherwise have to be solicited on each use of the system. It also allows the experience to be improved. Text-capable users' data includes, but need not be limited to: a. Preferred voice for delivery b. Membership of groups c. Billing plan d. Account information

[0069] In one embodiment, no User Preference data is stored for senders.

[0070] For receiving users such as Land-line Terminals (112) preferences include, but need not be limited to: a. Authentication mode (nil, self-authentication, PIN, Biomentric) b. Screening mode (none, screen all, screen unknown only) c. PIN number d. Biometric data for authentication of user e. Default Voice f. Default speed and volume settings g. Special voice parameters for known senders h. Recorded, or text names of known senders i. Whitelists j. Blacklists k. Silent hours 1. Alternate contacts and escalation rules

[0071] In one embodiment, there is no user preference data and the User Preference Store (122) can be omitted.

[0072] In one embodiment, there are many users associated with the same land-line Terminal (112). The User Preference Store (122) uses the Address (landline telephone number) to locate the details for a given user. They must therefore be further identified using an authentication mechanism. This is in turn used to identify the specific User Preference information from the otherwise identical records. This distinction may be used for differential billing if this is required.

[0073] Note that in the case that there are many users for a single non-text device, certain fields of the User Preference record must remain common to all instances. Any information that is required before the user is authenticated, including the type of authentication required, must be common to all users of a shared non-text device.

[0074] In one embodiment, the data common to all users at a shared address is extracted into a separate table.

[0075] Data is entered into, and retrieved from the User Preference Store (123) using System Configuration Terminals (150). In addition, the User Preference Store (123) is accessed and modified by the Message Delivery System (129).

[0076] The Text Message Store (124) holds text messages that are awaiting delivery. Each entry contains the full text data and all required delivery information and options.

The entries in this store are created by the Message Delivery System (129) after a message has been received from a text-capable device via a Text Messaging Gateway (160).

[0077] Each record in the Text Message Store (124) holds the time that the message is scheduled for delivery. When this time arrives, the Message Delivery System (129) retrieves the text message and attempts to deliver it according to the rules supplied and the preferences of the user to whom the message is addressed.

[0078] In one embodiment, in order to deliver text messages, the body of the message must be expanded to account for common text abbreviations. [0079| However, abbreviated spelling may be used in a less predictable way. In one embodiment, misspelled words are identified using a spell checker and tagged, such that during the delivery process they may be presented in their pronounced form or spelled to the user.

[0080] In one embodiment, a spell checker presents likely options that are carried with the message and offered as alternatives to the user during the dialog.

[0081] In one embodiment, a grammar checker is used in conjunction with each candidate identified by the spell checker in order to further prioritise the likely candidates.

[0082] The Recorded Media Store (128) holds data that may have been provisioned by the system owner as part of the dialog of delivery. This is an alternative to synthesized text and may be used for variety, to save computing power or cost. It also enables non-voice items such as music or tones to be played. It also holds data that is recorded by users themselves (such as user names) for use within in the dialog.

[0083] The Background Media Store (126) is largely the same as the Recorded Media Store (128), however, this data is only used to provide a backdrop or mood during the delivery process. The data is mixed, by summation, with the content being delivered.

[0084] The Message Delivery Log (121) is distinct from the Billing Information Store and is used for audit, diagnostics, monitoring and statistical purposes. It contains meaningful activity with respect to the components of the system. At a minimum each call attempt is recorded, however, it is desirable that all activity on the call is monitored. For performance reasons, the level of logging can be configured using a System Configuration Terminal (150).

[0085] The Message Delivery System (129) is responsible for the core behaviour of the system and is described in detail below.

[0086] With reference to Figure 2, Message Delivery System (129) can be further decomposed into a number of parts.

[0087] The Message Delivery Manager (210) is the core component and coordinates the activity of the system.

[0088] The Message Delivery Manager (210) receives text messages for delivery to a non-text terminal through a Text Message Gateway (160). The address of the non-text terminal is used to locate the User Preference information, and in turn the User Group information is located. This allows permissioning, billing and time of day information to be extracted and used in the subsequent processing of the message. The message is then stored in the Text Message Store (124). [0089) After a message that requires a response has been successfully delivered, the Message Delivery Manager (160) will send a response text message to a text-capable terminal via a Text Message Gateway (160). [0090) The Message Delivery Manager (210) is responsible for logging all information to the Message Delivery Log (121) and recording billing information to the Billing Information Store (125). [0091) When the message it about to be delivered to the non-text terminal, the Message Delivery Manager (210) instructs the Text Expander (220) to extract the text message from the Text Message Store (124) and tag items that should be pronounced in a manner other than the text would suggest. The expanded text is automatically forwarded to the TTS (230) engine for synthesis into renderable data. [0092) The Message Delivery Manager (210) also controls the operation of the TTS (230) engine. [0093) The Message Delivery Manager (210) controls the operation of Media Streamers (280) which take non-real-time media data delivers it onto a real time fabric for transmission. [0094) In one embodiment the data may be expanded and transformed in a manner suitable for FAX devices, should a FAX device be encountered. [0095) In one embodiment, the Message Delivery Manager (210) controls the Connection Fabric (240) to route data streams between the components connected to that fabric, as required during the delivery process. In another embodiment, the Connection Fabric (240) is passive and the interconnections are made by instructing the components to communicate with each other.

[0096] When required, the Message Delivery Manager (210) instructs a Media Streamers (280) to receive data from the Recorded Media Store (128). This is streamed to the end user in the place of TTS data.

[0097] When a background audio is required, the Message Delivery Manager (210) instructs a Media Streamers (280) to receive data from the Recorded Media Store (128) or an External Data Feed (130). In these cases, the Media Delivery Manager (210) must route the content and background streams to a Media Mixer (250) and instruct that mixer on how the media should be summed.

[0098] The Media Mixer (250) will perform the summing operation and emit a real time signal. The Media Delivery Manager (210) will ensure that this output is emitted to a Media Interface (270) that is connected to the call in progress.

[0099] The Message Delivery Manager (210) places calls to non-text terminals by instructing the Session Control (260) devices to perform the appropriate call signaling processes. If successful, this will result in a call and the media for that call will be received and transmitted on a Media Interface (270) device. By instructing the devices and/or the Connection Fabric (240), the media is streamed from the Media Interface (270) to a Signal Analyser (290), which determines the nature of the terminal device or agent operating that device. The Signal Analyser (290) may also be instructed to detect responses from the user during the progress of the dialog.

[00100] If the agent is non-automated and authentication is required for that address, then, depending on the type of authentication required, the Message Delivery Manager (210) will: a. Prompt the user for self-confirmation and instruct the Signal Analyser (290) to detect the digits, utterance or other response and then determine if this affirmative or otherwise. b. Prompt the user for a PIN and instruct the Signal Analyser (290) to detect the digits, utterance or other response, and then compare that with the PINs of the possible users. c. The Authenticator (295) is requested to perform analysis of the signal received in order to determine the identity of the user using the biomentric templates held in the User Preference store (122).

[00101] In one embodiment, an additional storage component (Advertising Message Store) holds advertising messages to be delivered and another component (Advertising Message Selector) is responsible for the intelligent selection of the advertising material to be delivered. In this embodiment, at the time that an advertising message is to be delivered, the Message Delivery Manager (210) requests an advertising message from the Advertising Message Selector). The Advertising Message Selector uses any number of rules and data in order to make the most appropriate selection. The rules and data may include, but need not be limited to: a. Time (relative, absolute, day, month year). b. Ambient conditions (eg weather) c. Location of receiving user d. History of past advertising delivery e. Preferences of the receiving user f. Identity of the sender g. Identity of the receiving user h. Billing regime of the receiving user i. The content of the text message being delivered j. Responses received from the user during the dialog

[00102] The level of interest shown by the user is stored in the Message Delivery Log (121) for reporting to the advertiser. This additional information may include, but need not be limited to: a. The identity of the message, b. Whether it was delivered completely, c. Whether the user requested to hear it again, d. Whether the user requested to hear more e. Whether the user requested to have information delivered to them at a later time, f. Whether the user requested, and was transferred to an automated agent, g. Whether the user requested, and was transferred to a live agent. h. A unique token to be used to correlate the call made by the system and that received by the live or automated agent to whom the user was connected.

[00103] With reference to Figure 3, the process of delivering a text message to a nontext terminal is coordinated by the Message Delivery Manager (210) and comprises the first steps of determining the next message to be delivered. This involves scanning the Text Message Store (124) for a message that is ready to be delivered, based on the timeframes supplied by the sender and the preferences of the users at the target address, as held in the User Preference Store (122).

[00104] On determining that a message is ready for delivery, the Message Delivery Manager (210) first examines the User Preference Store (122) and User Group Store (123) to determine whether blocking options apply. If the user allows anyone to send messages to them, then they may have entered a blacklist of forbidden users. If the user is part of a closed user group then anyone outside that group will be blocked.

[00105] If the sender is not allowed to send to the target user then the message is discarded. If the sender has requested a response then a response may be sent.

[00106] If the sender is not blocked then Message Delivery Manager (210) instructs a Session Control (260) device to place a call to the specified address.

[00107] If the call is not answered then the call is cleared and the status logged. Any special instructions within the text message sent by the sender regarding handling of failed calls are examined. The preferences of the users at the address being called are also examined. If it is determined that the call has failed then the text message is marked as failed, otherwise another call is scheduled, either to the same address or to another address.

[00108] In one embodiment, the sender's preferences take precedence over the user's.

[00109] In another embodiment, the user's preferences take precedence over the sender's.

[00110] If the call is answered, then the Signal Analyser (290) is used to determine the type of device or agent present at the receiving terminal. If a FAX is discovered, then the message is reformatted for delivery to a fax. User preferences are applied, for example, for security purposes, the message may only include reference to the fact that a caller has sent a message. The receiver may be asked to call another number and, optionally, enter a transaction code to retrieve that message. In other cases the user may wish to have the message delivered, in yet other cases, no message should be delivered.

[00111] The sender may also provide rules and even alternate messages to be delivered to FAX devices.

[00112] In one embodiment, the sender's preferences take precedence over the user's preferences in the case of a FAX.

[00113] In another embodiment, the user's preferences take precedence over the sender's.

[00114] Advertising material may be added to the image delivered to the FAX.

[00115] If the call is answered and the Signal Analyser (290) determines that the type of device or agent present is an automated message mailbox, then the message is reformatted for delivery to an automated message mailbox. User preferences are applied, for example, for security purposes, the message may only include reference to the fact that a caller has sent a message. The receiver may be asked to call another number and, optionally, enter a transaction code to retrieve that message. In other cases the user may wish to have the message delivered, in yet other cases, no message should be delivered.

[00116] The sender may also provide rules and even alternate messages to be delivered to automated mailbox devices.

[00117] In one embodiment, the sender's preferences take precedence over the user's in the case of automated mailbox devices.

[00118] In another embodiment, the user's preferences take precedence over the sender's preferences.

[00119] Advertising material may be added to the content delivered to the automated mailbox device.

[00120] If the call is answered and the Signal Analyser (290) determines that a human user is present, then the identity of the user is determined, to the degree that is required in the User Preference Store (122) or specified by the sender in the message itself. In one embodiment, the user's preference takes precedence. In another embodiment, the sender's preference takes precedence.

[00121] The appropriate authentication process is applied as described earlier. The user and sender preferences may require no authentication. In these cases the identity of the required user is assumed to have been successfully established.

[00122] If the call is answered, the Signal Analyser (290) determines that a human user is present and it is established that an acceptable user is not present then the message is reformatted for delivery to the "wrong user". User preferences are applied, for example, for security purposes, the message may only include reference to the fact that a caller has sent a message. The receiver may be asked to record a message for the intended target user, wherein that message requests that the target user calls another number and enter their PIN in order to retrieve the message. In other cases the user may wish to have the message delivered, in yet other cases, no message should be delivered.

[00123] The sender may also provide rules and even alternate messages to be delivered to the "wrong user".

[00124] In one embodiment, the sender's preferences take precedence over the user's in the case of the "wrong user" being contacted.

[00125] In another embodiment, the user's preferences take precedence over the sender's.

[00126] Advertising material may be added to the content delivered.

[00127] If the call is answered, the Signal Analyser (290) determines that a human user is present and it is established that user one that is to receive the message, then the message delivery dialog is executed.

[00128] The message dialog will announce the user before delivering the text message. The receiver may reject the call at that point.

[00129] If they reject the call and they may chose to tag that sender as permanently blocked.

[00130] If the sender is accepted then the message is delivered according to the preferences of the specific use that has been authenticated and the preferences of the sender.

[00131] In one embodiment, the sender's preferences take precedence over the user's in the case that the required user is contacted.

[00132] In another embodiment, the user's preferences take precedence over the sender's.

[00133] Advertising material may be added to the content delivered to authenticated user.

[00134] During the delivery of the text message, the user is offered many modes in order to assist the understanding of the message. The modes may be selected automatically in a sequence, the preferred mode or modes may be stored in the User Preference Store (122), or the modes are explicitly selected by the user. The modes include, but need not be limited to: a.

Repeating the message b.

Repeating part of the message or a single word c.

Speed d.

Volume e.

Voice f.

Spelling of all abbreviations g- Spelling of a specific word h.

Spelling of all words not in dictionary i.

Spelling of all words. j- Elimination of background audio

[00135] If the sender requests a response then the type of the response is determined and a dialog presented to the user during which the response is solicited. The response may be gathered using DTMF tones, voice recognition or any means other means available. The response is then sent to the sender via a Text Message Gateway (160).

[00136] If the user would like to communicate directly with the sender, and the system, the sender and the receiver permit this feature, then, by signaling to the system in the process of the dialog, the system will place a new call to the sender's number and connect the two parties in the case that the call is successful.

[00137] After or during the delivery of the text message, the receiver may determine that they do not want to continue and further, that future messages from that user should be blocked. If the sender is to be blocked then their address is added to the users blacklist in the User Preference Store (122).

[00138] In all cases where the call is not successfully delivered, the escalation process is executed. The escalation process involves determining if there are alternate numbers that the user has recorded in the User Preference Store (122). If there are alternate addresses, then depending on the nature of the failure, the next address may be used. The time between calls is determined by the system unless the user also has specified as preferences.

[00139] Typically, the case where the device is suspected or known to be attended (eg the 'busy' condition) is treated differently from the case where the device is not attended. This is because, in the case that the device is attended, there is a high likelihood of a success if a call is made after a short duration, and a lower chance of an alternate address being attended. The system will have default rules that can be overridden by the user preferences.

[00140] The sender may also provide alternate addresses to be called in the escalation process.

[00141] In one embodiment, the escalation preferences of the user take precedence over those of the sender. In another embodiment, the preferences of the sender take precedence over those of the user.

[00142] In one embodiment, the user's and sender's escalation preferences are 'exclusive', that is one or other of the escalation sequences is chosen. In another embodiment, the user's and sender's escalation preferences is 'inclusive', that is, both sequences are executed, but in order of precedence.

Claims (46)

CLAIMS We claim:

1. An intelligent messaging server for bridging text and nontext services comprising: a. At least one text message gateway configured to send text messages to at least one text terminal and receive text messages from at least one text terminal; b. At least one media interface configured to send real-time data to at least one real-time communications network and receive real-time data from at least one real time communications network; c. At least one session controller configured to control the establishment of at least one real-time media communications sessions between at least one real time media interface and at least one media terminal; d. At least one text-to-speech processor; e. At least one text message store configured to hold: i. Text messages awaiting delivery to users at media terminals, IPONZ - >; JUL 2006 -31 -- ii. Text messages recently delivered to users at media terminals wherein said text messages require responses to be delivered to the originating text terminal; f. A media store containing media files; g. At least one media streamer configured to stream media data in real time from at least one media data source to at least one real-time media interface, wherein said media data source is: i. A text-to-speech processor, ii. A media file held in a media store; h. At least one media analyser configured to determine the class of media terminal to which the system is connected; i. At least one signal analyser configured to detect and interpret responses supplied by the user at a media terminal; j. At least one activity log holding a record of significant activities and events during the processing of text messages. k. An automated message delivery manager configured to: IPONZ - h JUL 2006 -32- i. Control reception of text messages from at least one text message gateway, ii. Control scheduling of text messages for delivery to users at media terminals, iii. Control at least one session controller in order to establish a real-time media connection between media terminals and media interfaces, iv. Apply remedial action in the case of a failed attempt to establish a real-time media connection between media terminals and media interfaces, v. Control at least one signal analyser in order to determine the class of media terminal to which a media interface is connected and use the class of media terminal to select a mode of dialog to be delivered. vi. Control streaming of real-time data from at least one media data sources to at least one media interface, vii. Control at least one signal analyser in order to gather responses from the user at a media terminal, IPOWZ - 4 JUL 2006 33 - viii. Interpret said responses into readable text, ix. Control at least one text message gateway in order to deliver responses gathered from users at media terminals to the originator of a text message.

2. The intelligent messaging server of claim 1 further comprising at least one text expander configured to identify common abbreviations, misspellings and grammatical errors.

3. The intelligent messaging server of claim 1 further comprising at least one control connection to an automated call distributor wherein said control connection is used to pass information known about the user to at least one automated call distributor.

4. The intelligent messaging server of claim 1 further comprising a text-to-facsimile converter wherein the automated message delivery manager is configured to select a mode of dialog based on the class of media terminal from the following options: a. Deliver the complete text message as audio; b. Deliver an alternative message as audio; c. Deliver the complete message as a facsimile image; d. Deliver an alternative message as a facsimile image; IPONZ - « JUL 2006 e. Deliver no message and regard delivery as failed.

5. The intelligent messaging server of claim 1 wherein the automated message delivery manager is configured to select remedial action based on the results of a failed connection attempt wherein the remedial actions are: a. Record the failure and abandon the message delivery; b. Schedule a further connection attempt after a delay that is tuned for the case where the media terminal is busy; c. Schedule a further connection attempt after a delay that is tuned for the case where the user failed to respond to the connection request.

6. The intelligent messaging server of claim 1 further comprising at least one media mixer configured to overlay at least two media streams.

7. The intelligent messaging server of claim 6 wherein the media data store contains at least one selectable media data file suitable for mixing as an overlay to the dialog of the message delivery.

8. The intelligent messaging server of claim 1 wherein the media data store contains at least one selectable media data -35- IPONZ " « JUL 2006 file that is suitable for rendering as an introduction to the message content;

9. The intelligent messaging server of claim 1 wherein at least one selectable alternative delivery dialog is available;

10. The intelligent messaging server of claim 1 wherein at least one text-to-speech processor has at least one selectable alternative voice.

11. The intelligent messaging server of claim 1 wherein at least one text-to-speech processor has at least one selectable alternative speech prosody setting.

12. The intelligent messaging server of claim 1 further comprising a selectable media filter configured to modify the media of the message as it is delivered.

13. The intelligent messaging server of claim 1 further comprising a user preference store containing information specific to each text terminal user for use in customizing the delivery of messages sent by each text terminal user.

14. An intelligent messaging server as claimed in any one of the claims 7 to 12 wherein the selectable alternative is controlled through data previously supplied and stored in the user preference store.

15. An intelligent messaging server as claimed in any one of the claims 7 to 12 wherein the selectable alternative is -36- IPONZ - h JUL 2006 controlled by the sender through information included with the text message.

16. The intelligent messaging server of claim 1 wherein the name of the text message sender is supplied with the text and used in specific parts of the delivery dialog.

17. The intelligent messaging server of claim 13 wherein the name of the text message sender is held in the user preference store and used in specific parts of the delivery dialog.

18. The intelligent messaging server of claim 13 wherein the sender is able to submit text messages through at least one speed-dial address and each speed-dial address is mapped to a target user address using information stored in the user preference data.

19. An intelligent messaging server as claimed in any one of the preceding claims further comprising a user group store containing the identity of text message senders permissioned to send to a specific target media terminal.

20. An intelligent messaging server as claimed in any one of preceding claims wherein the media data store contains at least one advertising message and the dialog is configured to offer at least one level of detail. IPONZ -4 JUL 2006 -37-

21. The intelligent messaging server of claim 1 wherein the dialog presented to a media terminal user is configured to offer the option of being connected to a contact center agent.

22. The intelligent messaging server of claim 1 further comprising at least one external data feed wherein an additional function of the message delivery manager is to direct the data received from the external data feed to a media interface at specific times during the delivery dialog.

23. The intelligent messaging server of claim 1 further comprising at least one external data feed wherein an additional function of the message delivery manager is to direct the data received from the external data feed to a media mixer at specific times during the delivery dialog.

24. A method of delivering a text message to a user at a media terminal wherein said method comprises the steps of: a. Receiving a text message that includes the address of a target media terminal user over a data network; b. Storing said text message in the text message store; c. Attempting to establish a real-time communications session between a media interface and a media terminal identified by target address supplied with the text message; IPONZ -JUL 2006 -38- Upon detecting failure of the attempt to establish said real-time communications session, selecting a remedial action and skipping the steps e to i; Determining the media type; Upon determining that the media is facsimile machine, executing the sub-steps of: i. Creating a facsimile image using the full content of the text message, ii. Delivering said facsimile image to the media terminal, iii. Skipping steps g to i Upon determining that the media is an answer machine machine, executing the sub-steps of: i. Using a text-to-speech processor to create an alternative audio message using the full content text of the text message, ii. Delivering message to the media terminal, iii. Skipping steps h to i Upon determining that the media is audio, delivering the message using a method comprising the sub-steps: i. Expanding the text message using a text expander; IPONZ - 4 JUL 2006 -39- ii. Delivering a preamble dialog to the media terminal user; iii. Translating the message to audio using a text-to-speech processor and rendering it to the user; iv. Delivering the message content to the media terminal user; v. Detecting whether the message was understood and taking remedial action in the case that the message is not fully understood; vi. Determining if a response was requested by the originator of the text message and soliciting a response if required; vii. Delivering a farewell dialog; i. Upon successful delivery of the message, canceling the text message and returning any responses to the originating user.

25. The method of claim 24 wherein the remedial action taken in step d contains the further step of scheduling a retry at a later time to the same address;

26. The method of claim 25 wherein the retry time is determined based on whether the connection attempt failed IPONZ . 40 - -h JUL 2006 due to lack of response from the terminal user, a temporary error condition, or a permanent error condition.

27. The method of claim 24 wherein the remedial action taken in step d contains the further step of scheduling a retry at a later time to an alternative address;

28. The method of claim 24 wherein the remedial action taken in step d contains the further step tenninating the delivery of the message;

29. The method of claim 24 wherein the facsimile image created in step f does not include any part of the text of the message.

30. The method of claim 24 wherein the audio message created in step g does not include the any part of the text of the message.

31. The method of claim 24, wherein after detection of a successful connection to a real user, an authentication procedure is executed to establish the identity of the user at the terminal.

32. The method of claim 24 wherein the remedial action taken in step h upon determination that the message was not understood is to repeat the message. - 4i - IPONZ -« JUL 2006

33. The method of claim 24 wherein the remedial action taken in step h upon determination that the message was not understood is to repeat the message at a slower rate.

34. The method of claim 24 wherein the remedial action taken in step h upon determination that the message was not understood is to repeat the message with additional pauses between words.

35. The method of claim 24 wherein the remedial action taken in step h upon determination that the message was not understood is to repeat the message spelled out letter-by letter.

36. A method of delivering a text message to a user at a media terminal as claimed in any one of the claims 32 to 36 wherein the remedial action taken in step h is not effective and the process is repeated for further remedial actions until all have been attempted without success.

37. The method of delivering a text message to a user at a media terminal as claimed in any one of the claims 24 to 36, wherein each step and sub-step is recorded at a sufficient granularity for the purposes of auditing and customer billing.

38. The method of delivering a text message to a user at a media terminal as claimed in any one of the claims 24 to IPONZ " « JUL 2006 37, further comprising the step of analyzing the text message for question marks in order to determine whether a response is to be solicited from the user at a media terminal.

39. The intelligent messaging server of claim 1 further comprising at least one text processor configured to locatc text considered to be inappropriate and remove it prior to rendering the text.

40. The intelligent messaging server of claim 1 further comprising at least one text processor configured to locate text considered to be inappropriate and to replace it with a reference to a sound file prior to rendering the text.

41. The intelligent messaging server of claim 1 further comprising at least one text processor configured to locate text considered to be inappropriate and to replace it with a silence.

42. The intelligent messaging server of claim 1 further comprising at least one text processor configured to locate text considered to be inappropriate and to replace it with alternative text.

43. The intelligent messaging server of claim 1 wherein the session controller is further configured to control the establishment of at least one real-time media IPONZ - h JUL 2006 -43 - communications session between at least one real time media interface and at least one agent media terminal associated with a contact centre.

44. The intelligent messaging server of claim 1 wherein the automated message delivery manager is further configured to control at least one session controller in order to establish a real-time media connection between agent media terminals associated with a contact centre and media interfaces,

45. A system for delivering a text message to a user at a media terminal, substantially as herein described with reference to the accompanying figures.

46. A method of delivering a text message to a user at a media terminal, substantially as herein described with reference to the accompanying figures. IPONZ -1 JUL 2006 -44-