Intent discovery in audio or text-based conversation

Patent metadata

Field	Value
Publication number	US-9620147-B2
Application number	US-201514612989-A
Country	US
Kind code	B2
Filing date	Feb 3, 2015
Priority date	Jun 19, 2012
Publication date	Apr 11, 2017
Grant date	Apr 11, 2017

Abstract

Methods, systems, and computer program products for identifying one or more utterances that are likely to carry the intent of a speaker are provided herein. A method includes providing a transcript of utterances to a word weight scoring module to perform inverse document frequency based scoring on each word in the transcript, thereby generating a weight for each word; calculating a weight for each utterance in the transcript to generate weighted utterances by summing the weights or each constituent word in each utterance; comparing at least one weighted utterance to pre-existing example utterances carrying the intent of a speaker to determine a relevancy score for the at least one weighted utterance; and generating a ranked order of the at least one weighted utterance from highest to lowest intent relevancy score, wherein the highest intent relevancy score corresponds to the utterance which is most likely to carry intent of the speaker.

First claim

What is claimed is:

1. A method comprising:
providing at least one transcript of utterances from a conversation between two or more parties to a word weight scoring module to perform inverse document frequency based scoring on each word in the at least one transcript, thereby generating a weight for each word, wherein the inverse document frequency based scoring measures the frequency of each word throughout the at least one transcript;
calculating a weight for each utterance in the transcript to generate weighted utterances by assigning to each utterance the weight of the word with a maximum weight that occurs in each utterance;
comparing at least one weighted utterance to pre-existing example utterances carrying an intent of a speaker to determine a relevancy score for the at least one weighted utterance based on similarity to the example utterances; and
generating a ranked order of the at least one weighted utterance from highest to lowest intent relevancy score, wherein the highest intent relevancy score corresponds to the utterance which is most likely to carry intent of the speaker, and wherein said generating is carried out by a relevant propagation module executing on a hardware processor of a computing device.

2. The method of claim 1 , wherein said word weight scoring module includes a list of words likely indicative of speaker intent along with a weight associated with each word, wherein the weight characterizes the probability that the word is a word carrying the intent of the speaker.

3. The method of claim 1 , comprising:
performing term frequency-inverse document frequency based scoring on the weighted utterances.

4. The method of claim 1 , wherein the utterances are text-based utterances.

5. The method of claim 1 , wherein the similarity is based on a cosine similarity measure.

6. The method of claim 1 , comprising:
selecting the top N utterances from the ranked order to represent an intent summary of the conversation.

7. The method of claim 1 , comprising:
highlighting each utterance and/or region of the conversation that corresponds to the intent of the conversation.

8. The method of claim 7 , comprising:
displaying each highlighted utterance and/or region of the conversation that correspond to the intent of the conversation.

9. A computer program product, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computing device to cause the computing device to:
provide at least one transcript of utterances from a conversation between two or more parties to a word weight scoring module to perform inverse document frequency based scoring on each word in the at least one transcript, thereby generating a weight for each word, wherein the inverse document frequency based scoring measures the frequency of each word throughout the at least one transcript;
calculate a weight for each utterance in the transcript to generate weighted utterances by assigning to each utterance the weight of the word with a maximum weight that occurs in each utterance;
compare at least one weighted utterance to pre-existing example utterances carrying an intent of a speaker to determine a relevancy score for the at least one weighted utterance based on similarity to the example utterances; and
generate a ranked order of the at least one weighted utterance from highest to lowest intent relevancy score, wherein the highest intent relevancy score corresponds to the utterance which is most likely to carry intent of the speaker, and wherein said generating is carried out by a relevant propagation module executing on a hardware processor of the computing device.

10. The computer program product of claim 9 , wherein said word weight scoring module includes a list of words likely indicative of speaker intent along with a weight associated with each word, wherein the weight characterizes the probability that the word is a word carrying the intent of the speaker.

11. The computer program product of claim 9 , wherein the program instructions executable by a computing device further cause the computing device to:
perform term frequency-inverse document frequency based scoring on the weighted utterances.

12. The computer program product of claim 9 , wherein the utterances are text-based utterances.

13. The computer program product of claim 9 , wherein the similarity is based on a cosine similarity measure.

14. The computer program product of claim 9 , wherein the program instructions executable by a computing device further cause the computing device to:
select the top N utterances from the ranked order to represent an intent summary of the conversation.

15. The computer program product of claim 9 , wherein the program instructions executable by a computing device further cause the computing device to:
highlight each utterance and/or region of the conversation that corresponds to the intent of the conversation.

16. The computer program product of claim 15 , wherein the program instructions executable by a computing device further cause the computing device to:
display each highlighted utterance and/or region of the conversation that correspond to the intent of the conversation.

17. A system comprising:
a memory; and
at least one processor coupled to the memory and configured for:
providing at least one transcript of utterances from a conversation between two or more parties to a word weight scoring module to perform inverse document frequency based scoring on each word in the at least one transcript, thereby generating a weight for each word, wherein the inverse document frequency based scoring measures the frequency of each word throughout the at least one transcript;
calculating a weight for each utterance in the transcript to generate weighted utterances by assigning to each utterance the weight of the word with a maximum weight that occurs in each utterance;
comparing at least one weighted utterance to pre-existing example utterances carrying an intent of a speaker to determine a relevancy score for the at least one weighted utterance based on similarity to the example utterances; and
generating a ranked order of the at least one weighted utterance from highest to lowest intent relevancy score, wherein the highest intent relevancy score corresponds to the utterance which is most likely to carry intent of the speaker, and wherein said generating is carried out by a relevant propagation module executing on the at least one processor.

18. The system of claim 17 , wherein said word weight scoring module includes a list of words likely indicative of speaker intent along with a weight associated with each word, wherein the weight characterizes the probability that the word is a word carrying the intent of the speaker.

19. The system of claim 17 , wherein the at least one processor is configured for:
performing term frequency-inverse document frequency based scoring on the weighted utterances.

20. The system of claim 17 , wherein the at least one processor is configured for:
highlighting each utterance and/or region of the conversation that corresponds to the intent of the conversation; and
displaying each highlighted utterance and/or region of the conversation that correspond to the intent of the conversation.

Assignees

• IBM

Inventors

• Deshmukh Om D
• Joshi Sachindra
• Saurabh Saket
• Verma Ashish

Classifications

• G10L15/02

  Feature extraction for speech recognition; Selection of recognition unit · CPC title

• G10L21/10

  Transforming into visible information · CPC title

• G10L15/26

  Speech to text systems (G10L15/08 takes precedence) · CPC title

• G10L15/1822

  Parsing for meaning understanding · CPC title

• G10L15/18 Primary

  using natural language modelling · CPC title