AT&T Labs Research — Leading Invention, Driving Innovation

Systems, methods, and programs for generating an authorized profile for a text communication device or account, may sample a text communication generated by the text communication device or account during communication and may store the text sample. The systems, methods, and programs may extract a language pattern from the stored text sample and may create an authorized profile based on the language pattern. Systems, methods, and programs for detecting unauthorized use of a text communication device or account may sample a text communication generated by the device or account during communication, may extract a language pattern from the audio sample, and may compare extracted language pattern of the sample with an authorized user profile.

A computer implemented method includes accessing a digital image including a plurality of faces including a first face and a second face. The computer implemented method includes identifying a plurality of identification regions of the digital image including a first identification region associated with the first face and a second identification region associated with the second face. The computer implemented method also includes assigning the digital image to a first face cluster of a plurality of face clusters when a difference between data descriptive of the first identification region and data descriptive of a face cluster identification region of the first face cluster satisfies a threshold. The computer implemented method further includes assigning the digital image to a second face cluster of the plurality of face clusters based at least partially on a probability of the second face and the first face appearing together in an image.

A method and apparatus for providing a private broadcast channel service, e.g., a virtual closed circuit television service, on packet networks such as Service over Internet Protocol (SoIP) networks are disclosed. For example, a customer subscribes to the private broadcast channel service for a group having a plurality of confirmed members who have confirmed their desire to participate. The network service provider enables one or more of the group members to upload media content to the service provider's network. The service provider then broadcasts the uploaded media content to all the confirmed members such that each confirmed member may view the media content by simply tuning to the appropriate channel assigned for the private broadcast channel service.

Disclosed herein are systems, methods, and computer readable-media for adaptive media playback based on destination. The method for adaptive media playback comprises determining one or more destinations, collecting media content that is relevant to or describes the one or more destinations, assembling the media content into a program, and outputting the program. In various embodiments, media content may be advertising, consumer-generated, based on real-time events, based on a schedule, or assembled to fit within an estimated available time. Media content may be assembled using an adaptation engine that selects a plurality of media segments that fit in the estimated available time, orders the plurality of media segments, alters at least one of the plurality of media segments to fit the estimated available time, if necessary, and creates a playlist of selected media content containing the plurality of media segments.

This invention provides a network access system that permits communication device users to access a network such as the Internet, a cable network, a Digital Subscriber Line (DSL) network or a broadband network (which may include cable and DSL networks). A network access device receives commands from users and executes the commands such as send e-mail, search the network, log on to pages supported by other parties on the network, download information from the network, request information such as movies from the network, bidding in an auction, etc. A user profile saves preferences of a user for interacting with the network. The network access device may supply primitive commands for the user to use individually or to construct complex command macros so that by a single command, more complex functions may be performed using the network. By the above described techniques, users who do not have access to devices such as a personal computer may access the network and take advantage of the rich resources available through the network.

Disclosed herein are systems, methods, and computer readable-media for dynamically constructing audio in a video program. The method includes extracting video metadata from a video program displayed on a playback device to a viewer, extracting component metadata from a plurality of audio components stored in a media object library, extracting viewer preferences from a viewer profile, receiving synchronization information about the video program, identifying a segment of the video program susceptible to inserting an audio component based on extracted video metadata, component metadata, and viewer preferences, transmitting the audio component to the playback device and a set of instructions detailing how to insert the audio component in real time in the segment of the video program, and constructing audio in the video program at the playback device using the audio component and the set of instructions.

A method is disclosed that includes receiving a request to transmit a first multimedia file. The first multimedia file is associated with a first file descriptor. The method includes automatically generating a first generated content descriptor by analyzing the first multimedia file using a first set of analysis parameters. The method also includes comparing the first generated content descriptor to the first file descriptor and determining a first score based on a deviation of the first generated content descriptor from the first file descriptor.

A system and method is provided to automatically generate content for ITV products and services by processing primary media sources. In one embodiment of the invention, keywords are automatically extracted from the primary media sources using one or more of a variety of techniques directed to video, audio and/or textual content of the multimodal source. In some embodiments, keywords are then processed according to one or more disclosed algorithms to narrow the quantity of downstream processing that is necessary to associate secondary sources (reference items) with the primary video source. Embodiments of the invention also provide automatic searching methods for the identification of reference items based on the processed keywords in order to maximize the value added by the association of reference items to the video source.

A communications network includes a mobile communications device and a communications terminal. The device and terminal are engaged in a call which includes audio stream data. An operation is performed, either at the mobile device or in the supporting network, to determine a location of the mobile communications device and link metadata representing the determined mobile communications device location to the audio stream data sent from that mobile communications device for the call. At the communications terminal, the metadata is extracted from the audio stream data to allow for presentation of the location of the mobile communications device.

Systems and methods for using an annotation guide to label utterances and speech data with a call type. A method embodiment monitors labelers of speech data by presenting via a processor a test utterance to a labeler, receiving input from the labeler that selects a particular call type from a list of call types and determining via the processor if the labeler labeled the test utterance correctly. Based on the determining step, the method performs at least one of the following: revising the annotation guide, retraining the labeler or altering the test utterance.

The present invention is a system and method for enabling multicast synchronization of initially unicasted content. Multiple unicast streams are synchronized in order to convert the unicast streams into a multicast stream. Each unicast stream may be accelerated or slowed down in relation to a reference stream to a common point within each stream upon which the unicast streams are replaced by a multicast stream of the same content.

A subscriber to a remote call forwarding feature on a wire-line telephone line associated with a PBX (101) or central office (402), hereinafter, a "user", can enable that feature by entering a feature code from an IP terminal (112, 412). The user enters a feature code from his/her IP terminal (112, 412), which feature code is received by a remote server (for example, server 110) having access to an IP user directory (113, 413), hereinafter, for example, an active user registry (AUR) database, and forwards necessary information for invoking the feature in a determined manner to a Service Control Point (SCP) (106, 406). The SCP recognizes the feature code as a request to invoke the remote call forwarding feature on the user's wire-line telephone so that calls directed to that wire-line telephone are forwarded to, for example, the user's cellular phone. Receipt of that feature code by the SCP, identifies the cellular phone's Electronic Serial Number (ESN) from a provided cellular phone number or Mobile Identification Number (MIN) and identifies the user and the feature to be invoked for that user. By performing an SCP database (108, 408) lookup associated with that ESN and MIN, information is retrieved relating to what PBX (101) or switch (402) should be contacted to invoke the feature, how and where to contact that PBX or switch, and what specifically must be sent to that PBX or switch to invoke the feature on the user's telephone line.

A communications network includes a mobile communications device and a communications terminal. The device and terminal are engaged in a call which includes audio stream data. An operation is performed, either at the mobile device or in the supporting network, to determine a location of the mobile communications device and link metadata representing the determined mobile communications device location to the audio stream data sent from that mobile communications device for the call. At the communications terminal, the metadata is extracted from the audio stream data to allow for presentation of the location of the mobile communications device.

Systems and methods for monitoring labelers of speech data. To test or train labelers, a labeler is presented with utterances that have already been identified as belonging to a particular class or call type. The labeler is asked to assign a call type to the utterances. The performance of the labeler is measured by comparing the call types assigned by the labeler with the existing call types of the utterances. The performance of a labeler can also be monitored as the labeler labels speech data by occasionally having the labeler label an utterance that is already labeled and by storing the results.

A system and a method are provided for an online address directory that may be updated by the address owner in real time, via the Internet, the telephone, wireless web, or otherwise. The address entry may contain an address for the address owner, together with beginning and ending effective dates for the address to ensure that the address information is current for any given date. The address owner may select a level of security for the address entry, to permit open or restricted access to the address entry, and may specify those persons who may view the entry. A person seeking address information for the address owner may obtain that information by entering an identifier correlating to the address owner, the dates of interest, and if required, a key, into a query in the directory, thereby obtaining accurate address information for the date of interest.

This invention provides a network access system that permits communication device users to access a network such as the Internet, a cable network, a Digital Subscriber Line (DSL) network or a broadband network (which may include cable and DSL networks). A network access device receives commands from users and executes the commands such as send e-mail, search the network, log on to pages supported by other parties on the network, download information from the network, request information such as movies from the network, bidding in an auction, etc. A user profile saves preferences of a user for interacting with the network. The network access device may supply primitive commands for the user to use individually or to construct complex command macros so that by a single command, more complex functions may be performed using the network. By the above described techniques, users who do not have access to devices such as a personal computer may access the network and take advantage of the rich resources available through the network.

Systems and methods for generating an annotation guide. Speech data is organized and presented to a user. After the user selects some of the utterances in the speech data, the selected utterances are included in a class and/or call type. Additional utterances that belong to the class and/or call type can be found in the speech data using relevance feedback, data mining, data clustering, support vector machines, and the like. After a call type is complete, it is committed to the annotation guide. After all call types are completed, the annotation guide is generated.

The present invention is a system and method for enabling multicast synchronization of initially unicasted content. Multiple unicast streams are synchronized in order to convert the unicast streams into a multicast stream. Each unicast stream may be accelerated or slowed down in relation to a reference stream to a common point within each stream upon which the unicast streams are replaced by a multicast stream of the same content.

A method and apparatus for delivering a greeting from a group of individuals to an individual or another group of individuals is disclosed. Recipient information, contributor information, billing information, and an input from each contributor are received by the system. The system composes the greeting using the contributor inputs and delivers the greeting to the recipient in the form of an audio or multimedia greeting. All the contributors can simultaneously provide their input to the greeting or individual contributors can connect to the system (at any time before a pre-determined time limit) and add an input to the greeting after reviewing the prior inputs of other contributors.

A method and apparatus for sending messages to a portable device. The apparatus broadcasts query signals and receives an identification code from a portable device as a reply. The identification code can identify a portable device user, user preferences and portable device capabilities. Messages of various formats are retrieved from local and/or remote messaging systems based on the identification code. The apparatus can send the retrieved messages to the portable device and receive messages from the portable device.

A method of reducing transmission delays in a packet-switched telephone system. Each of the data packets carrying telephone voice data is under-filled when necessary in order to avoid undue delays in sending the packets to their designation. Under-filling of the packets allows a routine telephone conversation to take place with out being adversely affected by network delays or delays normally attributed to the time needed to accumulate sufficient data to fill a packet to its normal size. Packets may also be filled to their normal size by combining data from more than one telephone call if the data from the other calls is immediately available. Filling an otherwise under-filled packet with data from multiple telephone calls avoids the inefficiencies which would otherwise be present with under-filled packets for a single call.

A pseudo proxy server is provided for a host network when the host network experiences periods of congestion. The host network determines whether an arrival rate of data requests exceeds a predetermined threshold. If so, the host network adds an address of a pseudo proxy server to a list of servers aliased to the host network. Thereafter, data requests are routed to the pseudo proxy server on a round robin basis. The pseudo proxy server may be engaged without first loading information content from the host network on the pseudo proxy server.

A speech processor uses at least two speech processing features to enhance the quality of speech signals received by a user during a telephone call. The speech processing features are applied to the speech signals. However, the user only hears speech signals affected by one speech processing feature until both features have fully converged or ramped-up, and the two features are no longer interfering with each other. At that point, a flash-cut of the second speech processing feature is activated. The flash-cut instantaneously switches to speech signals affected by both features. This quick transition makes the speech processing features more noticeable to the user, and the user is not subjected to the period where the features interfere. Further, an optional audio indicator is generated before implementing the flash-cut, so the user is alerted to the flash-cut, and the speech processing features are even more noticeable.

A method for transferring money is described. This method can include providing a cash card having stored thereon an amount of money; making a telephone call to a predetermined telephone number using the cash card as a payment device for the telephone call; dialing a predetermined suffix to the predetermined telephone number followed by a desired dollar amount; debiting the cash card by the desired dollar amount; and, crediting an account associated with the predetermined telephone number by the desired dollar amount.

A method and apparatus can enter a call into a lottery, if the call is eligible. A group of individuals or other entities associated with the call for purposes of the lottery is identified; and if the call has won the lottery, at least one member of group is advised of the group's winning status. The odds of winning a lottery for the call can be adjusted based on a participation history for at least one member of the group. The participation history can include a past number of calls made or total call time logged by the caller, etc. Thus, for example, groups of frequent users of a communication service can be rewarded with increased odds of winning a lottery each time a request for communication services is made and communication services are provided.

A network interface device that allows a user to simultaneously receive/send a plurality of calls from/to a telecommunications network by multiplexing/demultiplexing the calls on a single telephone line. The network interface device allows for the provisioning of an extra line on demand to the user for incoming and outgoing purposes. The network interface device further selectively connects the calls with a plurality of devices in accordance with each devices availability and the stored preferences of the party receiving the call. When a call ends, the network interface device terminates the temporary virtual line assignment.

This invention provides a telephone-data network access system that permits voice communication device users to access a data network such as the Internet. A telephone-data network access device receives commands from callers and executes the commands on the data network such as send e-mail, search the data network, log on to pages supported by other parties on the data network, etc. A caller profile saves preferences of a caller for interacting with the data network. The telephone-data network access device may supply primitive commands for the caller to use individually or to construct complex command macros so that by a single command, more complex functions may be performed using the data network. By the above described techniques, callers who do not have access to devices such as a personal computer may access the data network and take advantage of the rich resources available through the data network.

An emergency breakthrough system allows users to break through a busy signal, using an automated rather than operator assisted technique. Upon receipt of a busy signal, rather than hanging up and dialing an operator to gain a manual assistance for third party call breakthrough, the calling party is able to keypad an enablement request, such as *85, requesting emergency call breakthrough. After that entry, the system awaits a breakthrough passcode, such as a four digit number, which if validated permits the user to breakthrough the busy signal and reach the called party, without operator intervention. Predesignated numbers can gain automated access by ANI identity of the calling number.

An individual transfers money from a cash card (or the account that the cash card is tied to) by making a wireless phone call to another individual. A combination of a MONDEX style card with the individual phone cards are used for this purpose. This electronic cash (ecash) ecash/phone card is inserted into the phone to activate it for billing use and serves as the basis for a cash transaction. For example, person A makes a telephone call to person B and dials a suffix after the number to indicate the transaction and the amount. For example, one could transfer money to another card or account holder from one's card by dialing their number followed by CASH#200# to transfer 200 dollars to the other account. The card is then capable of being used as a cash card to make purchases or as a phone card to make calls. A database links the identity of the caller to the cash account and the transaction appears as part of a bank statement or as part of a wireless phone bill. A user could use his wireless phone to dial an account number and transfer money by entering the amount via DTMF or by speaking the amount to an ASR application connected to the database. A user could also use the phone to make purchases via the card by just dialing a number and having an anonymous purchase made of the item. Since the user would be identified by the wireless network, the user would not have to enter any additional information about identification or credit card. The number that the user dialed could then debit the user's card directly and the item could be delivered without any need for the user to enter more information.

A method and a system for setting a priority level of a call in which a request is received for changing a priority level of a call to one of a plurality of available call priority levels, the priority level of the call is changed based on the received request, and the call is handled based on the changed priority level of the call. The request for changing the priority level of the call can be a request for a central office service feature for changing the priority of the call, receipt a call routed to a predetermined telephone number, such as an 800, an 888, a 900 or a charge-to-caller number, or receipt of a DTMF signal indicating the requested priority level. A message can be generated providing a menu for changing the priority level of the call having at least one available priority level with an associated charge rate for each available priority level. When the priority level of the call is changed, the call can be advanced in a call holding queue, a lower priority call can be interrupted or the call can be routed on a high-priority trunk capacity available for high-priority calls.

A subscriber to a remote call forwarding feature on a wire-line telephone line associated with a PBX (101) or central office (402) can enable that feature by dialing feature code from his cellular phone (103, 403). The subscriber dials a feature code from his cellular phone, which feature code is received by the cellular phone's Mobile Switching Center (MSC) (105, 405) and forwarded to a Service Control Point (SCP) (106, 406). The SCP recognizes the feature code as a request to invoke the remote call forwarding feature on the subscriber's wire-line telephone so that calls directed to that wire-line telephone are forwarded to the cellular phone. Receipt of that feature code by the SCP, together with the cellular phone's Electronic Serial Number (ESN) and Mobile Identification Number (MIN) identifies the subscriber and the feature to be invoked for that subscriber. By performing a database (108, 408) lookup associated with that ESN and MIN, information is retrieved relating to what PBX or switch should be contacted to invoke the feature, how and where to contact that PBX or switch, and what specifically must be sent to that PBX or switch to invoke the feature on the subscriber's telephone line.

A subscriber to alias telephone number e-mail system can retrieve both their voice-mail messages and e-mail messages by accessing only their e-mail system. Specifically, when a calling party calls the subscriber of an alias e-mail system who is unavailable to take the call, and leaves a voice-mail message on a network-based voice-mail system, the voice-mail system stores the message. The network-based voice-mail system then determines, through a network-based database retrieval, whether the called party is a subscriber to an alias telephone number e-mail system. In such a system, e-mail messages are addressed with a name portion including a subscriber's telephone number. A translation server then translates the alias telephone number e-mail address to the subscriber's actual e-mail address and forwards the message to the subscriber. If the called party is determined to be a subscriber to such a system, the voice-mail system converts the stored voice-mail message into a text or a .WAV file and includes the converted message in an e-mail message addressed to the called party/subscriber's alias telephone number e-mail address for transmission over the Internet. The translation server then converts the alias telephone number e-mail address to the subscriber's actual e-mail address and forwards the e-mail message to the called party/subscriber's e-mail repository server at that actual address for later retrieval by the called party.

A speech processor uses at least two speech processing features to enhance the quality of speech signals received by a user during a telephone call. The speech processing features are applied to the speech signals. However, the user only hears speech signals affected by one speech processing feature until both features have fully converged or ramped-up, and the two features are no longer interfering with each other. At that point, a flash-cut of the second speech processing feature is activated. The flash-cut instantaneously switches to speech signals affected by both features. This quick transition makes the speech processing features more noticeable to the user, and the user is not subjected to the period where the features interfere. Further, an optional audio indicator is generated before implementing the flash-cut, so the user is alerted to the flash-cut, and the speech processing features are even more noticeable.

A method and apparatus for rewarding communication service users can enter a call into a lottery, if the call is eligible. A party associated with the call for purposes of the lottery is identified; and if the call has won the lottery, the lottery winner is advised of the winning status. The odds of winning a lottery for the call can be adjusted based on a participation history for the associated party including a history of past lottery prizes won and refused. Thus, a lottery winner can be given the opportunity to refuse a lottery prize won in return for a chance to win a better prize in a future lottery. Prizes won by the lottery winner can also be matched to the lottery winner's preferences and/or the lottery winning call.

Some consumers may subscribe to several communications services, for example, a primary service that utilizes cable telephony and a secondary service that uses standard POTS over twisted pair. The disclosed apparatus allows the two services to work in tandem so that upon detecting loss of operation in one service, the telephones seamlessly switch to the other service. A media selection device is in functional parallel with an alternate local telephone network and an existing local telephone network. Upon detecting a loss of local power or cable telephony dial tone, the device switches to the local exchange network based on the assumption that the local exchange network is operable, despite the loss of local power. Alternatively the delay between off-hook and dial tone can be used to diagnose which of the two services are better able to complete the originating call. Incoming and outgoing calls are also selectably engageable on the local exchange network without disrupting telephone service on the alternate network. This selection can be manual or based upon other criteria such as time of day.

A speech processor uses a plurality of speech processing features to enhance the quality of speech signals received by a user during a telephone call. The activation of a second speech processing feature is delayed for a time duration measured after the initiation of a particular call or measured after the activation of a first speech processing feature. This delay eliminates any interference between the second speech processing feature and the first speech processing feature that is active when the call is initiated. The delay also increases the time it takes for the speech processing features to become fully effective, thus making the enhancement in the quality of the speech signals more noticeable to the user. Further, an optionally generated audio indicator is provided when the second speech processing feature is applied so that the additional feature is even more noticeable to the user.