Learning of dialogue states and language model of spoken information system

• US 6,839,671 B2
• Filed: 12/19/2000
• Issued: 01/04/2005
• Est. Priority Date: 12/20/1999
• Status: Expired due to Fees

• Current Assignee: British Telecommunications PLC (BT Group PLC)
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Attwater, David J., Edgington, Michael D., Durston, Peter J.
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): PATEL, KINARI M
• Application Number:
  US10/130,483
• Publication Number:
  US 20030091163A1
• Time in Patent Office: 1,477 Days
• Field of Search: 704/255, 704/257, 704/245, 704/236, 704/238, 704/200
• US Class Current: 704/255
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models

Abstract

In this invention dialogue states for a dialogue model are created using a training corpus of example human—human dialogues. Dialogue states are modelled at the turn level rather than at the move level, and the dialogue states are derived from the training corpus. The range of operator dialogue utterances is actually quite small in many services and therefore may be categorized into a set of predetermined meanings. This is an important assumption which is not true of general conversation, but is often true of conversations between telephone operators and people. Phrases are specified which have specific substitution and deletion penalties, for example the two phrases “I would like to” and “can I” may be specified as a possible substitution with low or zero penalty. Thus allows common equivalent phrases are given low substitution penalties. Insignificant phrases such as ‘erm’ are given low or zero deletion penalties.

Claims

1. A method of classifying a plurality of sequences of symbols to form a plurality of sets of sequences of symbols, the method comprising:

a) determining a distance between each sequence and each other sequence in said plurality of sequences in dependence upon a set of semantically insignificant symbol sequences and a set of equivalent symbol sequence pairs; and

b) grouping the plurality of sequences into a plurality of sets in dependence upon said distances;

wherein the symbols are words transcribed from call center operator speech signals spoken to a caller during an inquiry by the caller to the call center.

2. A method according to claim 1 wherein the words are transcribed from call center operator speech signals using a speaker dependent speech recogniser.

3. A method of generating a set of semantically insignificant symbol sequences for use in the method of claim 1, comprising:

classifying a plurality of sequences of symbols into a plurality of sets;

for each of the sets, determining an optimal alignment between each sequence thereof and each other sequence in that set; and

allocating a symbol or sequence of symbols to the set of semantically insignificant symbol sequences, the symbol or sequence of symbols having been deleted to obtain an optimal alignment between two sequences of a set.

4. A method of generating a set of equivalent symbol sequence pairs for use in the method of claim 1, comprising:

classifying a plurality of sequences of symbols into a plurality of sets;

determining an optimal alignment between each sequence in a set and each other sequence in that set; and

allocating a pair of symbols or sequences of symbols to the set of equivalent symbol sequences, the symbols or sequences of symbols having been substituted for each other to obtain an optimal alignment between two sequences of a set.

5. A method of generating a grammar for inquiries made to a call center, using the plurality of sets of sequences of words generated according to claim 1, comprising:

transcribing a plurality of inquiries according to which of the sets the sequences of words in the enquiry occur; and

generating a grammar in dependence upon the resulting transcription.

6. A method of measuring the occurrence of particular types of telephone inquiry received in a call center using the plurality of subsets of sequences of words generated according to claim 1.

12. A data carrier loadable into a computer and carrying instructions for causing the computer to carry out the method according to claim 1.

7. An apparatus for classifying a plurality of sequences of symbols to form a plurality of sets of sequences of symbols, the symbols being words transcribed from call center operator speech signals spoken to a caller during an inquiry from the caller to the call center, the apparatus comprising:

a store for storing a set of semantically insignificant symbol sequences;

a store for storing a set of equivalent symbol sequence pairs;

determining means connected to receive the transcribed call center operator speech signals and further arranged to determine a distance between each sequence and each other sequence in said plurality of sequences in dependence upon the set of semantically insignificant symbol sequences and the set of equivalent symbol sequence pairs; and

means for grouping the plurality of sequences into a plurality of sets in dependence upon said distances.

8. An apparatus according to claim 7, further comprising a speaker dependent recogniser for transcribing call center operator speech signals generated during the inquiry by the caller to the call center.

9. An apparatus for generating a set of semantically insignificant symbol sequences for use by the apparatus of claim 7, comprising:

a classifier for classifying a plurality of sequences of symbols into a plurality of sets;

alignment means for determining an optimal alignment for each of the sets between each sequence thereof and each other sequence in that set; and

means for allocating a symbol or sequence of symbols to the set of semantically insignificant symbol sequences, the symbol or sequence of symbols having been deleted to obtain an optimal alignment between two sequences of a set.

10. An apparatus for generating a set of equivalent symbol sequence pairs for use by the apparatus of claim 7, comprising:

a classifier for classifying a plurality of sequences of symbols into a plurality of subsets;

means for determining an optimal alignment between each sequence in a set and each other sequence in set; and

means for allocating a pair of symbols or sequences of symbols to the set of equivalent symbol sequences, the symbols or sequences of symbols having been substituted for each other to obtain an optimal alignment between two sequences of a set.

11. An apparatus for generating a grammar for inquiries made to a call center by a caller comprising:

a store for storing a plurality of sets of sequences of words, the sequences having been classified into the sets by an apparatus according to claim 7;

means for transcribing a plurality of inquiries according to which of the sets of sequences of words in the inquiry occur; and

means for generating a grammar in dependence upon the resulting transcription.

13. A data carrier loadable into a computer and carrying instructions for enabling the computer to provide the apparatus according to claim 7.

14. A method of classifying a plurality of sequences of words to form a plurality of sets of sequences of words, the method comprising:

transcribing the plurality of sequences of words from call center operator speech signals spoken to a caller during an inquiry by the caller to the call center;

determining a distance between each sequence of words and each other sequence of words in said plurality of sequences; and

grouping the plurality of sequences of words into a plurality of sets in dependence upon said distances.

15. A method according to claim 14, in which the words are transcribed from call center operator speech signals using a speaker dependent speech recogniser.

16. A method of generating a grammar for inquiries made by a caller to a call center, using the plurality of sets of sequences of words generated according to claim 14, comprising:

transcribing a plurality of inquiries according to which of the sets the sequences of words in the inquiry occur; and

generating a grammar in dependence upon the resulting transcription.

17. A method of measuring the occurrence of particular types of telephone inquiry received in a call center using the plurality of subsets of sequences of words generated according to claim 14.

18. An apparatus for classifying a plurality of sequences of words to form a plurality of sets of sequences of words, the apparatus comprising:

transcribing means for transcribing the plurality of sequences of words from call center operator speech signals spoken to a caller during an inquiry by the caller to the call center;

determining means connected to receive the transcribed call center operator speech signals and further arranged to determine a distance between each sequence and each other semantically insignificant symbol sequences and the set of equivalent symbol sequence pairs; and

means for grouping the plurality of sequences into a plurality of sets in dependence upon said distances.

19. An apparatus according to claim 18, wherein the transcribing means further comprise a speaker dependent recogniser for transcribing the call center operator speech signals generated during the inquiry from the caller to the call center.