Voice processing system

• US 6,282,268 B1
• Filed: 11/21/1997
• Issued: 08/28/2001
• Est. Priority Date: 05/06/1997
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Pickering, John Brian, Hulse, Brian, Jordan, Robert Michael, Ritchie, Andrew, Hughes, Jeremy Peter James, Maynard, Caroline Edith
• Primary Examiner(s): Hoosain, Allan
• Application Number:
  US08/976,201
• Time in Patent Office: 1,376 Days
• Field of Search: 379/67.1, 379/68, 379/88.01, 379/88.03, 379/88.08, 379/88.16, 379/88.02, 379/88.07, 379/88.17, 379/88.26, 379/201, 379/219, 379/242, 704/243, 704/254, 704/256, 704/260, 704/246
• US Class Current: 379/88.03
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
• CPC Class Codes: H04M 7/12 for working between exchanges having different types of switching equipment, e.g. power-driven and step by step or decimal and non-decimal

Abstract

A voice processing system comprising a digital trunk processor 135 and a system unit 145 incorporates a local voice recognition resource 185 provided on a digital signal processing board 180, and can access a remote voice recognition resource 320 in a server 300 via a local area network 250. The system provides a barge-in facility for the recognition resource on the remote server, whereby a prompt is played out to the user, and the incoming telephony signal is fed into a voice activity detector on the digital trunk processor. Responsive to a detection of incoming voice activity, the outgoing prompt is terminated, and the incoming data is transferred over the local area network to the remote server for recognition.

Claims

1. A voice processing system comprising:

a telephone interface unit over which telephony calls can be made and received, said telephone interface unit receiving an incoming audio signal as part of a telephone call;

an application program for controlling the processing of said telephony calls;

first means for connecting a local speech recognition resource on an adapter card to said telephone interface unit via a time division multiplex bus, for transferring said incoming audio signal from said telephone interface unit to said local speech recognition resource via a timeslot on said time division multiplex bus;

second means for connecting a remote speech recognition resource on a server system to said telephone interface unit via a local area network, for transferring said incoming audio signal from said telephone interface unit to said remote speech recognition resource in packets over said local area network;

wherein said local speech recognition resource is different from said remote speech recognition resource, the former being provided on a DSP within said voice processing system, the latter being a purely software system external to the voice processing system;

said voice processing system further comprising a voice technology subsystem providing an application interface for said application program to access said local or remote speech recognition resource to process said incoming audio signal to determine recognised text corresponding to said incoming audio signal, said application interface rendering it transparent to the application program whether the incoming audio signal is transferred from the telephone interface unit to the local speech recognition resource over said time division multiplex bus or to the remote speech recognition resource over said local area network.

2. The voice processing system of claim 1, wherein each speech recognition resource has an associated recognition server, and the voice technology subsystem includes means for receiving information from the recognition server concerning where the speech recognition resource is located.

3. The voice processing system claim 2, wherein said voice technology subsystem interacts with said recognition server via a predetermined recognition interface which is independent of where the speech recognition resource associated with said recognition server is located.

4. The voice processing system of claim 3, wherein the voice technology subsystem includes means for establishing a path for telephony data between the telephone interface unit and the speech recognition resource in accordance with the received information about where the speech recognition resource is located.

5. The voice processing system of claim 3, wherein said application program accesses said speech recognition resource via said application interface by specifying a named resource.

6. The voice processing system of claim 3, wherein said application program accesses said speech recognition resource via said application interface by specifying a desired recognition function.

7. The voice processing system of claim 2, wherein the voice technology subsystem includes means for establishing a path for telephony data between the telephone interface unit and the speech recognition resource in accordance with the received information about where the speech recognition resource is located.

8. The voice processing system of claim 4, wherein the voice technology subsystem includes a client component, which includes said means for establishing a telephony path, and a server component, which interacts with the recognition server, wherein the operation of said server component is independent of where the speech recognition resource is located.

9. The voice processing system of claim 8, wherein said application program accesses said speech recognition resource via said application interface by specifying a named resource.

10. The voice processing system of claim 8, wherein said application program accesses said speech recognition resource via said application interface by specifying a desired recognition function.

11. The voice processing system of claim 7, wherein the voice technology subsystem includes a client component, which includes said means for establishing a telephony path, and a server component, which interacts with the recognition server, wherein operation of said server component is independent of where the speech recognition resource is located.

12. The voice processing system of claim 7, wherein said application program accesses said speech recognition resource via said application interface by specifying a named resource.

13. The voice processing system of claim 7, wherein said application program accesses said speech recognition resource via said application interface by specifying a desired recognition function.

14. The voice processing system of claim 1, wherein said application program accesses said speech recognition resource via said application interface by specifying a named resource.

15. The voice processing system of claim 1, wherein said application program accesses said speech recognition resource via said application interface by specifying a desired recognition function.

16. A method of operating a voice processing system including a telephone interface unit, said method comprising the steps of:

making and receiving telephony calls over said telephone interface unit;

controlling the processing of said telephony calls with an application program;

transferring an incoming audio signal within a telephone call from said telephone interface unit to at least one of;

a local speech recognition resource connected to the telephone interface unit by a time division multiplex bus, via a timeslot on said time division multiplex bus; and

a remote speech recognition resource on a server system connected to said telephone interface unit via a local area network, in packets over said local area network;

wherein said local speech recognition resource is different from said remote speech recognition resource, the former being provided on a DSP within said voice processing system, the latter being a purely software system external to the voice processing system; and

providing an application interface for said application program to access a speech recognition resource to process an incoming audio signal from a telephony call to determine recognised text corresponding to said incoming audio signal, said application interface rendering it transparent to the application program whether the incoming audio signal is transferred from the telephone interface unit to the local speech recognition resource over said time division multiplex bus or to the remote speech recognition resource over said local area network.