Methods and apparatus for performing speaker independent recognition of commands in parallel with speaker dependent recognition of names, words or phrases

Abstract

Methods and apparatus for activating telephone services in response to speech are described. A directory including names is maintained for each customer. A speaker dependent speech template and a telephone number for each name, is maintained as part of each customer's directory. Speaker independent speech templates are used for recognizing commands. The present invention has the advantage of permitting a customer to place a call by speaking a person's name which serves as a destination identifier without having to speak an additional command or steering word to place the call. This is achieved by treating the receipt of a spoken name in the absence of a command as an implicit command to place a call. Explicit speaker independent commands are used to invoke features or services other than call placement. Speaker independent and speaker dependent speech recognition are performed on a customer's speech in parallel. An arbiter is used to decide which function or service should be performed when an apparent conflict arises as a result of both the speaker dependent and speaker independent speech recognition step outputs. Stochastic grammars, word spotting and / or out-of-vocabulary rejection are used as part of the speech recognition process to provide a user friendly interface which permits the use of spontaneous speech. Voice verification is performed on a selective basis where security is of concern.

Description

FIELD OF THE INVENTIONThe present invention is directed to telephone systems and, more particularly, to methods and apparatus for activating telephone services in response to speech.BACKGROUND OF THE INVENTIONTelephones are used to provide a host of services in addition to basic calling services. Such telephone services include services such as repeat dialing and call return where security is not of concern. Telephone services also include banking and financial services where security is of concern.Voice controlled dialing systems such as the one described in U.S. Pat. No. 5,165,095 permit a user to place a call verbally without knowing the number of the person being called. In accordance with the known system a user first speaks a command, e.g., the word "call" followed by a destination identifier. Once the command is identified using speaker independent voice recognition techniques, the system accesses speaker dependent or independent templates to recognize the destination identif...

AI Technical Summary

Benefits of technology

When a speaker independent command is detected, and no speaker independent name has been detected, the command is performed unless additional input is required, e.g., a name in the case of call forwarding, or security is of concern, such as in the case of message retrieval where the phone company is providing voice mail services. When security is of concern, a voice verification step is performed and the customer's identity is verified. Accordingly, the present invention provides a flexible system where voice verification is performed on an as needed basis and not necessarily on all calls.

Accordingly, the method and apparatus of the present invention permits a user to place a call by speaking a name without the need to first speak a steering word. The present invention also provides for call security via the selective application of voice verification in telephone transactions where security is of concern.

The use of stochastic grammars, word spotting and out of vocabulary word rejection features permit a customer to place a call or control the use of telephone services using language that is much closer to natural speech than is possible with other less flexible systems. For example, in accordance with the present invention the instruction <Call forwarding to Mary> can also be spoken as <Uhm . . . please . . . , I would like to activate call forwarding to . . . uhm . . . Mary, if I may>.

Problems solved by technology

Such format constraints require a user to speak in a manner that may be unnatural or uncomfortable for a user.

Images