Telephonic communications – Audio message storage – retrieval – or synthesis – Voice activation or recognition
Reexamination Certificate
1998-12-08
2001-05-08
{haeck over (S)}mits, T{overscore (a)}livaldis (Department: 2641)
Telephonic communications
Audio message storage, retrieval, or synthesis
Voice activation or recognition
C704S270000, C704S275000
Reexamination Certificate
active
06229881
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to speech recognition. More particularly, the present invention relates to a method and apparatus that provides enhanced speech recognition in a communication network.
BACKGROUND OF THE INVENTION
FIG. 1
is a block diagram of a known telephone call handling system that uses speech recognition. When a call is placed from a telephone
10
, a telephony signal is routed through a communication switch
20
. The communication switch
20
, such as a switch in the Public Switched Telephone Network (PSTN), can route the telephony signal to a communication network service provider operator platform
30
. In this case, a telephone operator can interact with the caller, such as to determine if the caller has a question about a telephone bill or needs directory assistance. Once the purpose of the call is determined, the operator can route the telephony signal as appropriate.
Having a live operator interact with callers can be expensive for the network service provider. Therefore, the service provider may instead route the telephony signal to an intelligent Voice Recognition Call Processing (VRCP) adjunct
100
. As used herein, an “adjunct” is a device that responds to a request for service processing in a communication network. As shown in
FIG. 1
, the VRCP adjunct
100
receives the telephony signal through a communication link, such as a PSTN digital T-1 connection, using media processing circuit boards.
An application infrastructure in the VRCP adjunct
100
automatically determines the purpose of the telephone call, such as by using speech recognition, and re-routes the telephony signal as required. If the VRCP adjunct
100
cannot determine the purpose of a particular call, the signal can be re-routed to the operator platform
30
. The Conversant Version 5.0 Personal Computer (PC) based system available from Lucent Technologies is an example of such a VRCP adjunct
100
.
A basic adjunct may simply respond to touch tone keypad signals, such as Dual-Tone Multi-Frequency (DTMF) signals, selected by the caller. For example, the adjunct may play the following message to a caller: “press ‘1’ if you have a billing question and press ‘2’for directory assistance.” The adjunct would then re-route the telephony signal based on the received DTMF signal.
The VRCP adjunct
100
shown in
FIG. 1
also has a limited amount of speech recognition capability. For example, the VRCP adjunct
100
may play the following message: “say the word ‘one’ if you have a billing question and say the word ‘two’ for directory assistance.” The VRCP adjunct
100
can then re-route the telephony signal if a “key word,” such as the word “one” or “two,” is recognized in the received speech signal. If the VRCP adjunct
100
does not recognize the received speech signal as being a key word, the telephony signal can be re-routed to the operator platform
30
. A live operator will then determine the purpose of the telephone call.
If the speech recognition capabilities of the VRCP adjunct
100
could be enhanced, such as by recognizing a larger number of key words or by recognizing commonly used phrases, the amount of operator time spent determining the purpose of telephone calls would be further reduced. The How May I Help You (HMIHY) system developed by AT&T is an example of a system with enhanced speech recognition capabilities.
There are a number of problems, however, with providing enhanced speech recognition capabilities in a communication network. For example, a number of VRCP adjuncts with limited capabilities may already be installed in the network. The hardware and/or software used in these existing, limited capability, VRCP adjuncts may be insufficient to support the more advanced processing techniques required by the enhanced speech recognition system, and it may be difficult to update the hardware used by a limited capability VRCP adjunct to provide enhanced speech recognition capabilities. For example, replacing and retesting existing limited capability VRCP adjuncts could be an expensive and time consuming task. Similarly, updating the software used by a limited capability VRCP may be prohibitively difficult.
In view of the foregoing, it can be appreciated that a substantial need exists for a method and apparatus that uses a limited capability VRCP adjunct to provide enhanced speech recognition capabilities in a communication network, and solves the other problems discussed above.
SUMMARY OF THE INVENTION
The disadvantages of the art are alleviated to a great extent by a method and apparatus that uses a limited capability VRCP adjunct to provide enhanced speech recognition capabilities in a communication network. In one embodiment of the present invention, a first input port is coupled to a first processor and receives a telephony signal from a communication network. A second input port, coupled to a second processor, receives the same telephony signal at substantially the same time as the telephony signal is received at first input port. The second processor generates recognized speech information based on the telephony signal, and a control line is coupled between the first and second processors. Based on the recognized speech information, the second processor sends a command to the first processor. The first processor then changes state, such as by re-routing the telephony signal, based on the command.
In another embodiment of the present invention, a first input port receives a telephony signal from a communication network, and a first processor coupled to the first input port enters one of a plurality of states based on the telephony signal. A second input port receives information associated with the telephony signal, and a second processor coupled to the second input port generates recognized speech information based on the received information associated with the telephony signal. The second processor also enters one of a plurality of states based on the information associated with the telephony signal such that the state of the second processor corresponds to the state of the first processor at a given point during the telephone call.
In still another embodiment of the present invention, a first input port is configured to receive a telephony signal from a communication network and a first processor, coupled to the first input port, generates speech data based on the telephony signal. A second input port coupled to a second processor receives the speech data from the first processor. The second processor generates recognized speech information based on this speech data. A control line coupled between said first and second processors lets the second processor send a command to the first processor based on the recognized speech information. The first processor may then re-route the telephony signal based on the command.
With these and other advantages and features of the invention that will become hereinafter apparent, the nature of the invention may be more clearly understood by reference to the following detailed description of the invention, the appended claims and to the several drawings attached herein.
REFERENCES:
patent: 5594789 (1997-01-01), Seazholtz et al.
patent: 5978450 (1999-11-01), McAllister et al.
patent: 6038305 (2000-03-01), McAllister et al.
patent: 6067347 (2000-05-01), Farris et al.
patent: 6104913 (2000-08-01), McAllister
William E. Longenbaker, et al. “Automation of Operator Services: A Successful Application of Speech Recognition Technology”, Proc. 2nd IEEE Workshop on Interactive Voice Technology for Telecommunications Applications (IVTTA 94), p. 161-164, Sep. 1994.*
R. J. Perdue, et al. “The Way We Were: Speech technology, Platforms, and Applications in the Old AT&T”, Proc. 3rd IEEE Workshop on Interactive Voice Technology for Telecommunication Applications (IVVTT 96), p. 7-11, Sep. 1996.
Alonso Tirso M.
Ammicht Egbert
Buntschuh Bruce Melvin
AT&T Corp
{haeck over (S)}mits T{overscore (a)}livaldis
LandOfFree
Method and apparatus to provide enhanced speech recognition... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus to provide enhanced speech recognition..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus to provide enhanced speech recognition... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2487975