Method and system for voice recognition menu navigation with error prevention and recovery

#The present invention relates in general to the field of voice recognition telephone services, and more specifically relates to a method and system for navigating a voice recognition menu with error prevention and recovery.

#Voice recognition technology has become increasingly popular for providing automated telephonic services through voice activation of menu selections. Voice activation of menu selections, such as for navigation through menu nodes for obtaining and providing information exchanges with automated services, offers a number of advantages over conventional touch tone DTMF signaling. For instance, with mobile phone networks, voice activation allows hands off menu selection for reduced distraction and improved safety of mobile phone use.

#One problem that often arises with the use of voice recognition over telephone networks is difficulty in accurately interpreting voice utterances by various users, especially where a user's possible responses may vary. One reason for interpretation difficulty relates to the nature of voice recognition, which compares an utterance from a user with a glossary of possible responses to determine the selection made by the user. Individuals tend to speak with a variety of accents so that a voice recognition system typically must have flexibility to recognize a range of utterance samples. Another reason for interpretation difficulty relates to unpredictable background noise that interferes with an utterance, such as a barking dog, a jack hammer or just a poor connection. The greater the variance of a situation from the situation that a voice recognition system is designed to handle, the greater the likelihood that voice recognition will fail and user utterances will not be recognized.

#The present invention may be better understood, and its numerous objects, features and advantages made apparent to those skilled in the art by referencing the accompanying drawings. The use of the same reference number throughout the several figures designates a like or similar element.

#FIG. 1 depicts a block diagram of a system for navigation of a voice recognition menu with error recovery;

#FIG. 2 depicts a flow diagram of a process for navigation of a voice recognition menu with error recovery; and

#Preferred embodiments of the present invention are illustrated in the figures, like numerals being used to refer to like and corresponding parts of the various drawings.

#To aid in the prevention of errors for voice activated menu navigation and in the recovery from errors that do occur, the present invention provides situation dependent voice activated navigation that accommodates a range of user preferences and behaviors to aid users in gracefully recovering from recognition or user errors. Thus, users are able to fluidly advance towards objectives in communication with a voice recognition system with reduced time and frustration. To accomplish these goals, in one embodiment, situation dependent utterance verification reduces error likelihood in an unobtrusive manner by applying different levels of verification for utterance interpretations based upon a determined confidence level for each interpretation. In another embodiment, a concise interrogative prompt obtains restatements of utterances having low confidence in a rapid manner and with minimal delay or introduction of confusion for the user. In another embodiment, dual modality provides DTMF signaling functionality, either on a selective basis to overcome difficult nodes or in combination with voice recognition to allow greater user flexibility.

#Referring now to FIG. 1, a block diagram depicts an error prevention and recovery system 10 for navigation of a voice recognition menu with reduced errors and improved recovery from errors that do occur. Users interface with error prevention and recovery system 10 through telephones 12 and a telephone network 14, such as the public switched telephone network or other telephonic communication systems like voice over internet protocol based systems. A voice recognition interactive voice response unit (IVR) 16 responds to user requests and guides users through a menu 18 having plural nodes for providing services, such as obtaining information from or giving information to users. Voice recognition IVR 16 accepts commands from users that are stated as utterances and converts the utterances into signals for use by a computer system to navigate through menu 18. For example, a mobile phone user may dial a voice activated menu to obtain the user's outstanding balance by stating “balance.” Voice recognition IVR 16 compares a digital sample of the user's utterance with sample utterances from a glossary 20 to determine that the user uttered “balance” and advances the user to the balance node of menu 18.