Data processing system for autonomously building speech identification and tagging data

• US 8,219,397 B2
• Filed: 06/10/2008
• Issued: 07/10/2012
• Est. Priority Date: 06/10/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Jaiswal, Peeyush, Khatri, Vikram S., Narayan, Naveen, Vialpando, Burt
• Primary Examiner(s): Smits, Talivaldis Ivars
• Assistant Examiner(s): PULLIAS, JESSE SCOTT
• Application Number:
  US12/136,342
• Publication Number:
  US 20090306979A1
• Time in Patent Office: 1,491 Days
• Field of Search: None
• US Class Current: 704/235
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models

Abstract

A method, system, and computer program product for autonomously transcribing and building tagging data of a conversation. A corpus processing agent monitors a conversation and utilizes a speech recognition agent to identify the spoken languages, speakers, and emotional patterns of speakers of the conversation. While monitoring the conversation, the corpus processing agent determines emotional patterns by monitoring voice modulation of the speakers and evaluating the context of the conversation. When the conversation is complete, the corpus processing agent determines synonyms and paraphrases of spoken words and phrases of the conversation taking into consideration any localized dialect of the speakers. Additionally, metadata of the conversation is created and stored in a link database, for comparison with other processed conversations. A corpus, a transcription of the conversation containing metadata links, is then created. The corpus processing agent also determines the frequency of spoken keywords and phrases and compiles a popularity index.

Claims

1. A method for autonomously creating a corpus of a conversation, the method comprising:

monitoring a conversation between one or more speakers;

identifying the spoken languages of the conversation;

identifying one or more topics being discussed within the conversation;

in response to identifying the topics being discussed, creating a plurality of metadata tags for each topic of the conversation, wherein the metadata tags include, for each topic of the conversation, one or more of;

a description of the speakers for a portion of the conversation,

a description of the languages spoken for a portion of the conversation,

a summary of the topic of the conversation for a portion of the conversation,

a plurality of links to other related topics of the conversation, and

a plurality of links to other related topics of a previously analyzed conversation;

storing the metadata tags in a link database;

determining a spoken emotional pattern of an autonomously selected topic of the conversation; and

creating a corpus of the conversation, wherein the corpus includes a text transcription of the conversation, and also includes an identification of the spoken emotional pattern and metadata tags of the conversation.

2. The method of claim 1, further comprising:

creating a paraphrase for each topic of the conversation; and

in response to determining the paraphrase, modifying the corpus to include the paraphrase.

3. The method of claim 2, further comprising:

referencing a lexical database to create a list of synonyms for words used in a paraphrase; and

in response to creating the list of synonyms, substituting the synonyms in the corpus.

4. The method of claim 1, further comprising, creating a popularity index of keywords and paraphrases of the conversation.

5. The method of claim 1, further comprising, determining a spoken emotional pattern of a speaker of the conversation for a selected topic of the conversation.

6. The method of claim 5, further comprising:

in response to determining the spoken emotional pattern, selecting a human emotion associated with the spoken emotional pattern; and

in response to selecting the human emotion, creating an identifier of the human emotion in the selected topic of the corpus.

7. A corpus processing agent comprising:

a processor;

a memory coupled to the processor;

an audio capture device; and

a processing logic for;

monitoring a conversation between one or more speakers;

identifying the spoken languages of the conversation;

identifying one or more topics being discussed within the conversation;

in response to identifying the topics being discussed, creating a plurality of metadata tags for each topic of the conversation, wherein the metadata tags include, for each topic of the conversation, one or more of;

a description of the speakers for a portion of the conversation,

a description of the languages spoken for a portion of the conversation,

a summary of the topic of the conversation for a portion of the conversation,

a plurality of links to other related topics of the conversation, and

a plurality of links to other related topics of a previously analyzed conversation;

storing the metadata tags in a link database;

determining a spoken emotional pattern of an autonomously selected topic of the conversation; and

creating a corpus of the conversation, wherein the corpus includes a text transcription of the conversation, and also includes an identification of the spoken emotional pattern and metadata tags of the conversation.

8. The corpus processing agent of claim 7, the processing logic further comprising:

creating a paraphrase for each topic of the conversation; and

in response to determining the paraphrase, modifying the corpus to include the paraphrase.

9. The corpus processing agent of claim 8, the processing logic further comprising:

referencing a lexical database to create a list of synonyms for words used in a paraphrase; and

in response to creating the list of synonyms, substituting the synonyms in the corpus.

10. The corpus processing agent of claim 7, the processing logic further comprising, creating a popularity index of keywords and paraphrases of the conversation.

11. The corpus processing agent of claim 7, the processing logic further comprising, determining a spoken emotional pattern of a speaker of the conversation for a selected topic of the conversation.

12. The corpus processing agent of claim 11, the processing logic further comprising:

in response to determining the spoken emotional pattern, selecting a human emotion associated with the spoken emotional pattern; and

in response to selecting the human emotion, creating an identifier of the human emotion in the selected topic of the corpus.

13. A computer-readable storage medium, not including transitory signals, having a plurality of instructions embodied therein, wherein the plurality of instructions, when executed by a processing device, causes a machine to:

monitor a conversation between one or more speakers;

identify the spoken languages of the conversation;

identify one or more topics being discussed within the conversation;

in response to identifying the topics being discussed, create a plurality of metadata tags for each topic of the conversation, wherein the metadata tags include, for each topic of the conversation, one or more of;

a description of the speakers for a portion of the conversation,

a description of the languages spoken for a portion of the conversation,

a summary of the topic of the conversation for a portion of the conversation,

a plurality of links to other related topics of the conversation, and

a plurality of links to other related topics of a previously analyzed conversation;

store the metadata tags in a link database;

determine a spoken emotional pattern of an autonomously selected topic of the conversation; and

create a corpus of the conversation, wherein the corpus includes a text transcription of the conversation, and also includes an identification of the spoken emotional pattern and metadata tags of the conversation.

14. The computer-readable storage medium of claim 13, the plurality of instructions further comprising instructions for enabling the machine to:

create a paraphrase for each topic of the conversation; and

in response to determining the paraphrase, modify the corpus to include the paraphrase.

15. The computer-readable storage medium of claim 14, the plurality of instructions further comprising instructions for enabling the machine to:

reference a lexical database to create a list of synonyms for words used in a paraphrase; and

in response to creating the list of synonyms, substitute the synonyms in the corpus.

16. The computer-readable storage medium of claim 13, the plurality of instructions further comprising instructions for enabling the machine to, create a popularity index of keywords and paraphrases of the conversation.

17. The computer-readable storage medium of claim 13, the plurality of instructions further comprising instructions for enabling the machine to, determine a spoken emotional pattern of a speaker of the conversation for a selected topic of the conversation.

18. The computer-readable storage medium of claim 17, the plurality of instructions further comprising instructions for enabling the machine to:

in response to determining the spoken emotional pattern, select a human emotion associated with the spoken emotional pattern; and

in response to selecting the human emotion, create an identifier of the human emotion in the selected topic of the corpus.