Abstract: A computer-implemented method for dynamically generating a speech recognition grammar is provided. The method includes determining a target entry, and accessing a plurality of potential distracters. The method also includes selecting one or more distracters from the plurality of potential distracters. More particularly, each potential distracter selected is selected based upon an assessed acoustic dissimilarity between the distracter and the target entry. The method further includes dynamically generating a speech recognition grammar that includes the target entry and one or more of the distracters selected based upon an acoustic dissimilarity to the target entry.

Abstract: In one embodiment of the invention, a method for selecting hold music corresponding to an expected waiting time comprises establishing a communications link with a caller; marking a first event by placing the caller on hold; determining a waiting time between the first event and a second event; selecting a media file having a parameter corresponding to the determined waiting time; playing the media file over the established communications link to the caller; and re-determining a waiting time remaining until the second event, selecting a different media file having a parameter corresponding to the re-determined waiting time, and playing the different media file over the established communications link to the caller.

Abstract: A data base is created for storage of voice prints of requesters that are believed to be fraudulently attempting to access information for which they are not authorized to obtain. When a user opens an account a voice print is obtained and stored. The user also provides answers to security related questions. When a requester tries to access the information, they must be authenticated by providing a voice print and answers to the security questions. If the voice print and answers do not result in a satisfactory match based on predetermined criteria, access is denied and the voice print is stored as a possibly fraudulent voice print. Subsequent access attempts are compared to the stored possibly fraudulent voice print which is reclassified as a likely fraudulent voice print if matched. Thus, unauthorized access is less likely.

Abstract: A computer-implemented method for dynamically generating a speech recognition grammar is provided. The method includes determining a target entry, and accessing a plurality of potential distractors. The method also includes selecting one or more distractors from the plurality of potential distractors. More particularly, each potential distractor selected is selected based upon an assessed acoustic dissimilarity between the distractor and the target entry. The method further includes dynamically generating a speech recognition grammar that includes the target entry and one or more of the distractors selected based upon an acoustic dissimilarity to the target entry.

Abstract: A wizard that from a fixed design can create various audio interfaces. The generated interfaces can be speech only, DTMF only, or various mixed speech and DTMF UIs. When specifying both speech and DTMF prompts, a number of combinations of these interfaces could be automatically generated. Robust speech recognition systems can be built by automatically generating a “shadow” DTMF application. The DTMF application will perform the same task as the primary speech application; however the transfer to a DTMF application could be done explicitly by the user, or could be transferred automatically (either a temporary or permanent transition) at a point in the call flow where there was a problem with the speech recognition.

Abstract: In a personal communications device having a touch sensitive display screen, text may be marked by a user to be used with different applications. In particular, to begin marking, a user has to maintain his touch at a location on the screen for a short predetermined period of time until the system detects that the position is held over the same character for a delayed period, at which time the system beeps to provide an indication that there has been a change of operation mode. The marking mode continues in operation until the user removes his contact from the touch screen. Thus, text over which the contact point passes after the initiation of the marking mode is marked, until the user removes his contact point from the touch screen. A second beep is sounded to alert the user of the termination of the marking mode, and the appearance of a menu of operations on the screen. The menu of operations echoes the marked text for user confirmation.

Abstract: A data processing device and method for searching a database having a plurality of entries to locate entries which match a search string. Each entry within the database is arranged into one or more predetermined search fields which each contain a character string. In response to receipt of a search string, a determination is made whether the search string is a subset of a particular entry within the database. In response to the determination that the search string is a subset of a particular entry, the particular entry is recorded as a matching entry. However, in response to a determination that the search string is not a subset of a particular entry, a determination is made whether a modified search string formed by removing delimiters from the search string is a subset of a character string within a particular predetermined search field within the particular entry.