← Back to Patent Explorer
US5675707A Expired / Lapsed US ⏱️ 0.0 years remaining

Automated call router system and method

Assignee: Nuance Communications Inc
•
Filing: 1995-09-15
•
Granted: 1997-10-07
•
Expires: 2015-09-15
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract An automated call routing system and method which operates on a call-routing objective of a calling party expressed in natural speech of the calling party. The system incorporates a speech recognition function, as to which a calling party's natural-speech call routing objective provides an input, and which is trained to recognize a plurality of meaningful phrases, each such phrase being related to a specific call routing objective. Upon recognition of one or more of such meaningful phrases in a calling party's input speech, an interpretation function then acts on such calling party's routing objective request to either implement the calling party's requested routing objective or to enter into a dialog with the calling party to obtain additional information from which a sufficient confidence level can be attained to implement that routing objective.

Inventors

  • 👤 Allen Louis Gorin
  • 👤 Barry Allen Parker
  • 👤 James B. Scherer
  • 👤 Jay Gordon Wilpon

Bibliographic Details

Application Number —
Jurisdiction US
Data Source google_patents
Last Synchronized 2026-10-31 00:19
CPC Classifications
G; G06; G06F; G06F40/00; G06F40/20; G06F40/279; G06Q; G06Q30/00; G06Q30/06; G06F40/205; G06F40/216; G06F40/289; G10; G10L; G10L15/00; G10L15/08; G10L15/18; G10L15/1822

Claims Analysis

Total Claims: 48
Claim 1
Independent Claim 129 words
An automated task classification system which operates on a task objective of a user, said task objective being expressed in natural speech of said user, comprising: a meaningful phrase generator that determines a plurality of meaningful phrases from a set of speech utterances, each of said meaningful phrases being selected based on one of a predetermined set of said task objectives; input recognizing means for recognizing at least one of said meaningful phrases in input speech of said user, said input recognizing means having as an input said meaningful phrases determined by said meaningful phrase generator; and interpretation means responsive to an input of said recognized meaningful phrases for making a classification decision based on said recognized meaningful phrases as to one of said set of predetermined task objectives.
Claim 2
Dependent Claim 129 words
An automated task classification system which operates on a task objective of a user, said task objective being expressed in natural speech of said user, comprising: a meaningful phrase generator that determines a plurality of meaningful phrases from a set of speech utterances, each of said meaningful phrases being selected based on one of a predetermined set of said task objectives; input recognizing means for recognizing at least one of said meaningful phrases in input speech of said user, said input recognizing means having as an input said meaningful phrases determined by said meaningful phrase generator; and interpretation means responsive to an input of said recognized meaningful phrases for making a classification decision based on said recognized meaningful phrases as to one of said set of predetermined task objectives.
Claim 3
Dependent Claim 29 words
The automated task classification system of claim 1 wherein said interpretation means further includes a dialog means for enabling said system to obtain a feedback response from said user.
Claim 4
Dependent Claim 23 words
The automated task classification system of claim 2 wherein said feedback response includes additional information with respect to said user's initial input speech.
Claim 5
Dependent Claim 31 words
The automated task classification system of claim 2 wherein said feedback response includes confirmation with respect to at least one of said set of task objectives determined in said classification decision.
Claim 6
Dependent Claim 44 words
The automated task classification system of claim 1 wherein said relationship between said determined meaningful phrases and said predetermined set of task objectives includes a measure of usefulness of a one of said meaningful phrases to a specified one of said predetermined task objectives.
Claim 7
Dependent Claim 18 words
The automated task classification system of claim 5 wherein said usefulness measure is manifested as a salience measure.
Claim 8
Dependent Claim 56 words
The automated task classification system of claim 6 wherein said salience measure is represented as a conditional probability of said task objective being requested given an appearance of said meaningful phrases in said input speech, said conditional probability being a highest value in a distribution of said conditional probabilities over said set of predetermined task objectives.
Claim 9
Dependent Claim 30 words
The automated task classification system of claim 6 wherein each of said plurality of meaningful phrases determined by said means for determining has a salience measure exceeding a predetermined threshold.
Claim 10
Dependent Claim 35 words
The automated task classification system of claim 1 wherein said relationship between said determined meaningful phrases and said predetermined set of task objectives includes a measure of commonality within a language of said meaningful phrases.
Claim 11
Dependent Claim 19 words
The automated task classification system of claim 9 wherein said commonality measure is manifested as a mutual information measure.
Claim 12
Dependent Claim 31 words
The automated task classification system of claim 10 wherein each of said plurality of meaningful phrases determined by said means for determining has a mutual information measure exceeding a predetermined threshold.
Claim 13
Dependent Claim 22 words
The automated task classification system of claim 1 wherein said interpretation means includes a confidence function in respect to said classification decision.
Claim 14
Dependent Claim 29 words
The automated task classification system of claim 1 wherein said input speech from said user represents a request for a specified one of said set of predetermined task objectives.
Claim 15
Dependent Claim 28 words
The automated task classification system of claim 1 wherein said input speech is responsive to a query from said system of a form "How may I help you?".
Claim 16
Dependent Claim 41 words
The automated task classification system of claim 1 wherein each of said speech utterances are directed to one of said set of predetermined task objectives and each said utterance is labelled with said one task objective to which it is directed.
Claim 17
Dependent Claim 132 words
An automated call routing system which operates on a routing objective of a calling party, said routing objective being expressed in natural speech of said calling party, comprising: a meaningful phrase generator that determines a plurality of meaningful phrases from a set of speech utterances, each of said meaningful phrases being selected based on one of a predetermined set of said routing objectives; input recognizing means for recognizing at least one of said meaningful phrases in input speech of said calling party, said input recognizing means having as an input said meaningful phrases determined by said meaningful phrase generator; and interpretation means responsive to an input of said recognized meaningful phrases for making a classification decision based on said recognized meaningful phrases as to one of said set of predetermined routing objectives.
Claim 18
Dependent Claim 132 words
An automated call routing system which operates on a routing objective of a calling party, said routing objective being expressed in natural speech of said calling party, comprising: a meaningful phrase generator that determines a plurality of meaningful phrases from a set of speech utterances, each of said meaningful phrases being selected based on one of a predetermined set of said routing objectives; input recognizing means for recognizing at least one of said meaningful phrases in input speech of said calling party, said input recognizing means having as an input said meaningful phrases determined by said meaningful phrase generator; and interpretation means responsive to an input of said recognized meaningful phrases for making a classification decision based on said recognized meaningful phrases as to one of said set of predetermined routing objectives.
Claim 19
Dependent Claim 30 words
The automated call routing system of claim 16 wherein said interpretation means further includes a dialog means for enabling said system to obtain a feedback response from said calling party.
Claim 20
Dependent Claim 24 words
The automated call routing system of claim 17 wherein said feedback response includes additional information with respect to said calling party's initial input speech.
Claim 21
Dependent Claim 31 words
The automated call routing system of claim 17 wherein said feedback response includes confirmation with respect to at least one of said set of routing objectives determined in said classification decision.
Claim 22
Dependent Claim 44 words
The automated call routing system of claim 16 wherein said relationship between said determined meaningful phrases and said predetermined set of routing objectives includes a measure of usefulness of a one of said meaningful phrases to a specified one of said predetermined routing objectives.
Claim 23
Dependent Claim 18 words
The automated call routing system of claim 20 wherein said usefulness measure is manifested as a salience measure.
Claim 24
Dependent Claim 56 words
The automated call muting system of claim 21 wherein said salience measure is represented as a conditional probability of said routing objective being requested given an appearance of said speech fragment in said input speech, said conditional probability being a highest value in a distribution of said conditional probabilities over said set of predetermined muting objectives.
Claim 25
Dependent Claim 30 words
The automated call routing system of claim 21 wherein each of said plurality of meaningful phrases determined by said means for determining has a salience measure exceeding a predetermined threshold.
Claim 26
Dependent Claim 35 words
The automated call routing system of claim 16 wherein said relationship between said determined meaningful phrases and said predetermined set of routing objectives includes a measure of commonality within a language of said meaningful phrases.
Claim 27
Dependent Claim 19 words
The automated call routing system of claim 24 wherein said commonality measure is manifested as a mutual information measure.
Claim 28
Dependent Claim 31 words
The automated call routing system of claim 25 wherein each of said plurality of meaningful phrases determined by said means for determining has a mutual information measure exceeding a predetermined threshold.
Claim 29
Dependent Claim 22 words
The automated call routing system of claim 16 wherein said interpretation means includes a confidence function in respect to said classification decision.
Claim 30
Dependent Claim 29 words
The automated call routing system of claim 16 wherein said input speech from said user represents a request for a specified one of said set of predetermined routing objectives.
Claim 31
Dependent Claim 28 words
The automated call routing system of claim 16 wherein said input speech is responsive to a query from said system of the form "How may I help you?".
Claim 32
Dependent Claim 41 words
The automated call routing system of claim 16 wherein each of said speech utterances are directed to one of said set of predetermined routing objectives and each said utterance is labelled with said one routing objective to which it is directed.
Claim 33
Independent Claim 107 words
A method for automated task classification which operates on a task objective of a user, said task objective being expressed in natural speech of said user, comprising the steps of: determining a plurality of meaningful phrases from a set of speech utterances, each of said meaningful phrases being selected based on one of a predetermined set of said task objectives; receiving as inputs said determined meaningful phrases and recognizing at least one of said meaningful phrases in input speech of said user; and making a classification decision in response to an input of said recognized meaningful phrases as to one of said set of predetermined task objectives.
Claim 34
Independent Claim 107 words
A method for automated task classification which operates on a task objective of a user, said task objective being expressed in natural speech of said user, comprising the steps of: determining a plurality of meaningful phrases from a set of speech utterances, each of said meaningful phrases being selected based on one of a predetermined set of said task objectives; receiving as inputs said determined meaningful phrases and recognizing at least one of said meaningful phrases in input speech of said user; and making a classification decision in response to an input of said recognized meaningful phrases as to one of said set of predetermined task objectives.
Claim 35
Dependent Claim 28 words
The automated task classification method of claim 31 including the additional step of entering into a dialog with said user to obtain a feedback response from said user.
Claim 36
Dependent Claim 23 words
The automated task classification method of claim 32 wherein said feedback response includes additional information with respect to said user's initial input speech.
Claim 37
Dependent Claim 31 words
The automated task classification method of claim 32 wherein said feedback response includes confirmation with respect to at least one of said set of task objectives determined in said classification decision.
Claim 38
Dependent Claim 44 words
The automated task classification method of claim 31 wherein said relationship between said determined meaningful phrases and said predetermined set of task objectives includes a measure of usefulness of a one of said meaningful phrases to a specified one of said predetermined task objectives.
Claim 39
Dependent Claim 18 words
The automated task classification method of claim 35 wherein said usefulness measure is manifested as a salience measure.
Claim 40
Dependent Claim 56 words
The automated task classification system of claim 36 wherein said salience measure is represented as a conditional probability of said task objective being requested given an appearance of said meaningful phrase in said input speech, said conditional probability being a highest value in a distribution of said conditional probabilities over said set of predetermined task objectives.
Claim 41
Dependent Claim 29 words
The automated task classification method of claim 36 wherein each of said plurality of meaningful phrases determined by said determining step has a salience measure exceeding a predetermined threshold.
Claim 42
Dependent Claim 35 words
The automated task classification method of claim 31 wherein said relationship between said determined meaningful phrases and said predetermined set of task objectives includes a measure of commonality within a language of said meaningful phrases.
Claim 43
Dependent Claim 19 words
The automated task classification method of claim 39 wherein said commonality measure is manifested as a mutual information measure.
Claim 44
Dependent Claim 29 words
The automated task classification method of claim 40 wherein each of said plurality of meaningful phrases determined by determining step has a mutual information measure exceeding a predetermined threshold.
Claim 45
Dependent Claim 20 words
The automated task classification method of claim 31 wherein said step of making a classification decision includes a confidence function.
Claim 46
Dependent Claim 29 words
The automated task classification method of claim 31 wherein said input speech from said user represents a request for a specified one of said set of predetermined task objectives.
Claim 47
Dependent Claim 25 words
The automated task classification method of claim 31 wherein said input speech is responsive to a query of a form "How may I help you?".
Claim 48
Dependent Claim 41 words
The automated task classification method of claim 31 wherein each of said speech utterances are directed to one of said set of predetermined task objectives and each said utterance is labelled with said one task objective to which it is directed.

References Cited (Backward Citations)

US4777600A Cited
US4866778A Cited
US5029214A Cited
US5033088A Cited
US5384892A Cited
US5390279A Cited
US5434906A Cited

USPTO Annuity & Maintenance Windows

Annuity Stage Window Opens Payment Deadline Surcharge Period Until Estimated Fee Status
3.5 Years (Year 4) 2000-10-06 2001-04-07 2001-10-07 $2000.00 Window Closed / Lapsed
7.5 Years (Year 8) 2004-10-06 2005-04-07 2005-10-07 $3760.00 Window Closed / Lapsed
11.5 Years (Year 12) 2008-10-06 2009-04-07 2009-10-07 $7700.00 Window Closed / Lapsed

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag