Synchronizing distributed speech recognition

• US 9,208,785 B2
• Filed: 05/10/2006
• Issued: 12/08/2015
• Est. Priority Date: 05/10/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Ben-David, Shay, Cross, Charles W. Jr.
• Primary Examiner(s): Shah, Paras D
• Application Number:
  US11/382,573
• Publication Number:
  US 20070265851A1
• Time in Patent Office: 3,499 Days
• Field of Search: 704/270, 704/275, 704/270.1, 704/231
• US Class Current: 1/1
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

Methods, apparatus, and computer program products are disclosed for synchronizing distributed speech recognition (‘DSR’) that include receiving in a DSR client notification from a voice server of readiness to conduct speech recognition and, responsive to the receiving, transmitting by the DSR client, from the DSR client to the voice server, speech for recognition.

Claims

1. A method of synchronizing speech recognition, the method comprising:

receiving, in a client terminal having access to a voice channel and a data channel distinct from the voice channel, an audio signal including speech of a user;

processing the audio signal to generate a processed speech signal, the processing comprising filtering the audio signal to remove one or more non-speech portions of the audio signal;

receiving notification from a voice server of readiness to conduct speech recognition;

responsive to the notification of readiness to conduct speech recognition, transmitting the processed speech signal over the data channel from the client terminal to the voice server for recognition;

receiving implicit notification from the voice server to cease transmitting speech, by completion of an X+V event or a VoiceXML event associated with a prompt and a transmission of user speech; and

responsive to the implicit notification from the voice server to cease transmitting speech, stopping speech transmission.

2. The method of claim 1, wherein the notification of readiness to conduct speech recognition is received through an application level data communications protocol.

3. The method of claim 1, further comprising discarding, by the client terminal, speech received prior to signaling the user to speak.

4. The method of claim 1, further comprising:

receiving audio signals;

detecting, by the client terminal, among the received audio signals, voice activity of the user; and

identifying only audio signals received during the voice activity as speech for recognition to be transmitted to the voice server.

5. The method of claim 1, further comprising:

providing to the user a speech prompt advising the user of information to be provided; and

responsive to receiving the notification of readiness, signalling the user to speak.

6. The method of claim 1, wherein the processed speech signal is transmitted using Voice Over Internet Protocol (VOIP).

7. The method of claim 1, wherein the processed speech signal is transmitted over a mobile packet-switched data channel to the voice server for recognition.

8. The method of claim 1, wherein the processed speech signal is transmitted over a cellular packet-switched data channel to the voice server for recognition.

9. An apparatus for synchronizing speech recognition, the apparatus comprising:

a communications adapter having access to a voice channel and a data channel distinct from the voice channel;

at least one computer processor; and

a computer memory operatively coupled to the at least one computer processor, the computer memory having disposed within it computer program instructions that, when executed by the at least one computer processor, perform a method comprising;

receiving an audio signal including speech of a user;

processing the audio signal to generate a processed speech signal, the processing comprising filtering the audio signal to remove one or more non-speech portions of the audio signal;

receiving notification from a voice server of readiness to conduct speech recognition;

responsive to the notification of readiness to conduct speech recognition, transmitting the processed speech signal over the data channel to the voice server for recognition;

receiving implicit notification from the voice server to cease transmitting speech, by completion of an X+V event or a VoiceXML event associated with a prompt and a transmission of user speech; and

responsive to the implicit notification from the voice server to cease transmitting speech, stopping speech transmission.

10. The apparatus of claim 9, wherein the notification of readiness to conduct speech recognition is received through an application level data communications protocol.

11. The apparatus of claim 9, wherein the method further comprises discarding speech received prior to signaling the user to speak.

12. The apparatus of claim 9, wherein the method further comprises:

receiving audio signals;

detecting, among the received audio signals, voice activity of the user; and

identifying only audio signals received during the voice activity as speech for recognition to be transmitted to the voice server.

13. The apparatus of claim 9, wherein the method further comprises:

providing to the user a speech prompt advising the user of information to be provided; and

responsive to receiving the notification of readiness, signalling the user to speak.

14. The apparatus of claim 9, wherein the processed speech signal is transmitted using Voice Over Internet Protocol (VOIP).

15. The apparatus of claim 9, wherein the processed speech signal is transmitted over a mobile packet-switched data channel to the voice server for recognition.

16. The apparatus of claim 9, wherein the processed speech signal is transmitted over a cellular packet-switched data channel to the voice server for recognition.

17. A computer program product comprising:

at least one non-transitory recordable medium encoded with instructions that, when executed, perform a method for synchronizing speech recognition, the method comprising;

receiving, in an apparatus having access to a voice channel and a data channel distinct from the voice channel, an audio signal including speech of a user;

processing the audio signal to generate a processed speech signal, the processing comprising filtering the audio signal to remove one or more non-speech portions of the audio signal;

receiving notification from a voice server of readiness to conduct speech recognition;

responsive to the notification of readiness to conduct speech recognition, transmitting the processed speech signal over the data channel to the voice server for recognition;

receiving implicit notification from the voice server to cease transmitting speech, by completion of an X+V event or a VoiceXML event associated with a prompt and a transmission of user speech; and

responsive to the implicit notification from the voice server to cease transmitting speech, stopping speech transmission.

18. The computer program product of claim 17, wherein the notification of readiness to conduct speech recognition is received through an application level data communications protocol.

19. The computer program product of claim 17, wherein the method further comprises discarding speech received prior to signaling the user to speak.

20. The computer program product of claim 17, wherein the method further comprises:

receiving audio signals;

detecting, among the received audio signals, voice activity of the user; and

identifying only audio signals received during the voice activity as speech for recognition to be transmitted to the voice server.

21. The computer program product of claim 17, wherein the method further comprises:

providing to the user a speech prompt advising the user of information to be provided; and

responsive to receiving the notification of readiness, signalling the user to speak.

22. The computer program product of claim 17, wherein the processed speech signal is transmitted using Voice Over Internet Protocol (VOIP).

23. The computer program product of claim 17, wherein the processed speech signal is transmitted over a mobile packet-switched data channel to the voice server for recognition.

24. The computer program product of claim 17, wherein the processed speech signal is transmitted over a cellular packet-switched data channel to the voice server for recognition.