Computer-implemented system and method for performing distributed speech recognition

• US 10,121,475 B2
• Filed: 10/31/2016
• Issued: 11/06/2018
• Est. Priority Date: 03/29/2007
• Status: Active Grant

• Current Assignee: Intellisist, Inc. dba Spoken Communications (Avaya Incorporated)
• Original Assignee: Intellisist, Inc. dba Spoken Communications (Avaya Incorporated)
• Inventors: Odinak, Gilad
• Primary Examiner(s): Neway, Samuel G
• Application Number:
  US15/339,788
• Publication Number:
  US 20170047070A1
• Time in Patent Office: 736 Days
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/193 Formal grammars, e.g. finite state automata, context free grammars or word networks
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 15/32 Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems
• CPC Class Codes: G10L 15/34 Adaptation of a single recogniser for parallel processing, e.g. by use of multiple processors or cloud computing
• CPC Class Codes: H04M 2201/38 Displays
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 3/51 Centralised call answering arrangements requiring operator intervention , e.g. call or contact centers for telemarketing
• CPC Class Codes: H04M 3/5166 in combination with interactive voice response systems or voice portals, e.g. as front-ends
• CPC Class Codes: H04M 3/5183 Call or contact centers with computer-telephony arrangements

Abstract

A computer-implemented system and method for performing distributed speech recognition is provided. Audio data is collected. A main grammar and secondary grammars are simultaneously provided for the audio data. Each secondary grammar includes an independent grammar. Speech recognition is simultaneously performed on the audio data using each secondary grammar. A new grammar is constructed for the audio data based on the main grammar template using results of the speech recognition. Further speech recognition is performed on the audio data using the new grammar.

Claims

1. A computer-implemented system for performing distributed speech recognition, comprising:

audio data;

a grammar module to simultaneously provide a main grammar to a main recognizer and secondary grammars for the audio data to a plurality of secondary recognizers, wherein each secondary grammar comprises an independent grammar and the main recognizer is interfaced to each of the secondary recognizers;

a speech recognition module to simultaneously perform speech recognition on the audio data using each secondary grammar;

a new grammar module to construct a new grammar for the audio data based on results of the speech recognition from each of the secondary grammars and a structure of the main grammar;

the main recognizer to perform further speech recognition on the audio data using the new grammar; and

a processor to execute the modules.

2. A system according to claim 1, further comprising:

a prompt module to transmit prompts to a telephony interface, wherein the prompts comprise call information provided in at least one of a file and script; and

a receipt module to receive from a caller the audio data in reply to one or more of the transmitted prompts.

3. A system according to claim 2, further comprising:

a selection module to select at least one of the prompts for playback to the caller.

4. A system according to claim 2, further comprising at least one of:

a playback module to automatically play the prompt to the caller when the file comprises speech; and

a conversion module to convert the prompt into speech for playback to the caller when the prompt comprises text.

5. A system according to claim 1, further comprising:

a receipt module to receive from an agent one or more of a further prompt and modifications to results of the further speech recognition.

6. A system according to claim 1, further comprising:

a console to display results of the further speech recognition, wherein the console is managed by an agent.

7. A system according to claim 1, wherein the independent grammars comprise language-specific grammars.

8. A system according to claim 1, further comprising:

an assignment module to assign a confidence level to results of the further speech recognition; and

a threshold module to apply a predetermined threshold to the confidence level.

9. A system according to claim 8, further comprising:

a placeholder module to provide a placeholder with the results of the further speech recognition when the confidence level of the results falls below the threshold, wherein the placeholder provides notification that further results are being generated.

10. A system according to claim 1, further comprising:

secondary recognizers to perform the speech recognition, wherein each secondary recognizer utilizes one of the independent grammars identified via a secondary grammar reference.

11. A computer-implemented method for performing distributed speech recognition, comprising:

collecting audio data;

simultaneously providing a main grammar to a main recognizer and secondary grammars for the audio data to a plurality of secondary recognizers, wherein each secondary grammar comprises an independent grammar and the main recognizer is interfaced to each of the secondary recognizers;

simultaneously performing speech recognition on the audio data using each secondary grammar;

constructing a new grammar for the audio data based on results of the speech recognition from each of the secondary grammars and a structure of the main grammar; and

performing further speech recognition on the audio data using the new grammar.

12. A method according to claim 11, further comprising:

transmitting prompts to a telephony interface, wherein the prompts comprise call information provided in at least one of a file and script; and

receiving from a caller the audio data in reply to one or more of the transmitted prompts.

13. A method according to claim 12, further comprising:

selecting at least one of the prompts for playback to the caller.

14. A method according to claim 12, further comprising at least one of:

automatically playing the prompt to the caller when the file comprises speech; and

converting the prompt into speech for playback to the caller when the prompt comprises text.

15. A method according to claim 11, further comprising:

receiving from an agent one or more of a further prompt and modifications to results of the further speech recognition.

16. A method according to claim 11, further comprising:

displaying results of the further speech recognition on a console managed by an agent.

17. A method according to claim 11, wherein the independent grammars comprise language-specific grammars.

18. A method according to claim 11, further comprising:

assigning a confidence level to results of the further speech recognition; and

applying a predetermined threshold to the confidence level.

19. A method according to claim 18, further comprising:

providing a placeholder with the results of the further speech recognition when the confidence level of the results falls below the threshold, wherein the placeholder provides notification that further results are being generated.

20. A method according to claim 11, further comprising:

performing the speech recognition via secondary recognizers, wherein each secondary recognizer utilizes one of the independent grammars identified via a secondary grammar reference.