Conversational networking via transport, coding and control conversational protocols

• US 20020184373A1
• Filed: 03/21/2002
• Published: 12/05/2002
• Est. Priority Date: 11/01/2000
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Maes, Stephane H.
• Granted Patent: US 6,934,756 B2
• Time in Patent Office: Days
• US Class Current: 709/228
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: H04L 65/1101 Session protocols
• CPC Class Codes: H04L 65/1104 Session initiation protocol [SIP]
• CPC Class Codes: H04L 65/1106 Call signalling protocols; H.323 and related
• CPC Class Codes: H04L 65/65 Network streaming protocols, e.g. real-time transport protocol [RTP] or real-time control protocol [RTCP]
• CPC Class Codes: H04L 65/70 Media network packetisation
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 69/32 Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04M 3/4938 comprising a voice browser which renders and interprets, e.g. VoiceXML
• CPC Class Codes: H04N 21/6437 Real-time Transport Protocol [RTP]
• CPC Class Codes: H04N 21/8106 involving special audio data, e.g. different tracks for different languages

Abstract

A system and method for implementing conversational protocols for distributed conversational networking architectures and/or distributed conversational applications, as well as real-time conversational computing between network-connected pervasive computing devices and/or servers over a computer network. The implementation of distributed conversational systems/applications according to the present invention is based, in part, on a suitably defined conversational coding, transport and control protocols. The control protocols include session control protocols, protocols for exchanging of speech meta-information, and speech engine remote control protocols.

Claims

1. A DSR (distributed speech recognition) communication stack, comprising:

a session control layer for managing a communication session, negotiating an upstream and downstream codec at initiation of the session, dynamically switching the upstream or downstream codec during the session, managing an uplink data transmission comprising DSR encoded data, and for managing a downlink transmission comprising results of server-side speech processing; and

a transport control layer comprising RTP (real-time protocol), or extensions thereof, for transmitting DSR encoded data and RTCP (real time control protocol), or extensions thereof, for controlling transmission of the DSR encoded data.

2. The communication stack of claim 1, wherein the session control layer supports (i) SDP (session description protocol) over SIP (session initiation protocol), or extensions thereof, or (ii) SOAP (simple object access protocol) over SIP, or extensions thereof.

3. The communication stack of claim 2, further comprising a transport layer that supports UDP or an extension thereof.

4. The communication stack of claim 2, further comprising a transport layer that supports TCP (transmission control protocol) or an extension thereof.

5. The communication stack of claim 1, wherein the session control layer implements H.323 or an extension thereof.

6. The communication stack of claim 1, wherein the communication stack exchanges speech meta-information in band using RTP.

7. The communication stack of claim 1, wherein the communication stack exchanges speech meta-information out of band using the session control layer.

8. The communication stack of claim 7, wherein the session control layer supports SIP.

9. The communication stack of claim 1, wherein the communication stack to exchanges speech meta-information out of band using RTCP or an extension thereof.

10. The communication stack of claim 1, wherein the communication stack exchanges speech meta-information via HTTP (hypertext transfer protocol) or an extension thereof.

11. The communication stack of claim 1, wherein different DSR encoded data are transmitted on separate RTP streams.

12. The communication stack of claim 1, wherein different DSR encoded data streams are transmitted sequentially in one RTP stream separated by a frame marker.

13. The communication stack of claim 1, further comprising a mechanism for specifying when a DSR encoded data stream is to be transmitted with guaranteed delivery.

14. The communication stack of claim 1, wherein the session control layer supports barge-in.

15. The communication stack of claim 1, further comprising an engine control layer for remote control of conversational engines.

16. The communication stack of claim 15, wherein the engine control layer exchanges meta information to determine engine capabilities and to reserve engines.

17. The communication stack of claim 6, wherein the engine control layer supports RTSP (real time streaming protocol) or an extension thereof.

18. The communication stack of claim 16, wherein the engine control layer supports WSDL (web services description language) or an extension thereof.

19. The communication stack of claim 18, wherein the engine control layer supports WSDL or an extension thereof over SOAP or an extension thereof.

20. The communication stack of claim 16, wherein the engine control layer supports SOAP (simple object access protocol) over RTSP.

21. A method for providing network communication, comprising the steps of:

establishing a network connection;

negotiating an initial uplink codec and an initial downlink codec;

determining conversational engine capabilities;

transmitting DSR (distributed speech recognition) data encoded with a selected uplink codec scheme, using a real time protocol;

transmitting engine control data to remotely control an engine for processing the DSR data; and

dynamically negotiating and switching between different uplink and/or downlink codecs during a communication session in response to a predetermined request.