Speech recognition system

Claim 3

A speech recognition system including the system for creating word models recited in Claim 1, characterised in that: -said speech recognition system further includes; means for making an acoustic description of a given portion of speech to be recognised; -means for temporarily storing said acoustic description; -means for storing acoustic word models for each of a plurality of words; -means for storing a sequence of textual characters in association with each acoustic word model; and -recognition means for selecting which one or more of said acoustic models best match said acoustic description and for producing a list of those best matching acoustic models; said means for indicating a menu including means for providing the user with a menu of the character sequences associated with the best matching acoustic models on the list produced by the recognition means; and -said means for making an acoustic model including means for using as one of the utterances used in the making of said acoustic model said acoustic description used by said recognition means in selecting the character sequences indicated in said menu.

Claim 4

-3. A speech recognition system including the system for creating word models recited in Claim 1, characterised in that: -said system further includes; -means for making an acoustic description of a menu selection command to be recognized; -means for storing acoustic models of a plurality of menu selection commands each associated with one of the sequences of textual characters indicated by said menu; and -recognition means for selecting which of said menu selection command models best matches said acoustic description of said menu selection command to be recognized; and -said means for enabling the user to select a given character sequence from the menu includes means for using said means for making an acoustic description of said menu selection command to be recognized, said means for storing menu selection command models, and said recognition means to respond to the user's speaking of a menu selection command by selecting the character sequence from the menu corresponding to the best matching menu selection command model selected by said recognition means.

Claim 5

-4. A speech recognition system including the system for creating word models recited in Claim 1, characterised in that: -said system further includes -means for making an acoustic description of an editing command to be recognized; -means for storing acoustic models of a plurality of editing commands each associated with a function for editing a sequence of textual characters; -recognition means for selecting which one or more of said editing command models best matches said acoustic description of said editing command to be recognized; -said means for enabling the user to edit the selected character sequence includes means for using said means for making an acoustic description of said editing command to be recognized, said means for storing editing command models, and said recognition means to respond to the user's speaking of an editing command by performing upon the selected character sequence the editing function corresponding to the best matching editing command model selected by said recognition means.

Claim 6

-5. A speech recognition system which includes the system for creating word models described in claim 1 characterised in that, said speech recognition system further includes -means for representing a body of test comprised of one or more words and for representing a word insertion location relative to said text; -recognition means for recognizing a spoken word by selecting a word which matches said spoken word; and -means for inserting a representation of either the word selected by said recognition means or the character sequence selected by the user into said body of text at said word insertion location.

Claim 7

-6. An adaptive speech recognition method for recognizing a plurality of spoken words over a period of time and for improving a set of acoustic word models used during that recognition, said method comprising the steps of: -storing a set of such acoustic word models, with each such model being stored in association with a word label; -forming an acoustic description of the sound of each spoken word to be recognized; -attempting to perform automatic speech recognition upon each such acoustic description by comparing it against a plurality of said acoustic word models to select which one or more of said word models best match it; -storing each of certain acoustic descriptions in association with a word label corresponding to the spoken word represented by said acoustic description, and, for each such certain acoustic descriptions, performing this storing after attempting to perform such recognition upon it and before attempting to perform such recognition upon others of said acoustic descriptions; -associating a given acoustic description with a word label and updating the acoustic word model of that label by seeking to find any acoustic descriptions previously stored in association with that word label and by merging acoustic data from those previously stored descriptions with the acoustic data from the given acoustic description to make the updated word model, and performing this storing after attempting to perform such recognition upon said given acoustic description and before attempting to perform such recognition upon others of said acoustic descriptions; -storing the updated acoustic word model in said set of word models and using it in the subsequent performance of said attempted automatic speech recognition.

Claim 8

-6. An adaptive speech recognition method for recognizing a plurality of spoken words over a period of time and for improving a set of acoustic word models used during that recognition, said method comprising the steps of: -storing a set of such acoustic word models, with each such model being stored in association with a word label; -forming an acoustic description of the sound of each spoken word to be recognized; -attempting to perform automatic speech recognition upon each such acoustic description by comparing it against a plurality of said acoustic word models to select which one or more of said word models best match it; -storing each of certain acoustic descriptions in association with a word label corresponding to the spoken word represented by said acoustic description, and, for each such certain acoustic descriptions, performing this storing after attempting to perform such recognition upon it and before attempting to perform such recognition upon others of said acoustic descriptions; -associating a given acoustic description with a word label and updating the acoustic word model of that label by seeking to find any acoustic descriptions previously stored in association with that word label and by merging acoustic data from those previously stored descriptions with the acoustic data from the given acoustic description to make the updated word model, and performing this storing after attempting to perform such recognition upon said given acoustic description and before attempting to perform such recognition upon others of said acoustic descriptions; -storing the updated acoustic word model in said set of word models and using it in the subsequent performance of said attempted automatic speech recognition.

Claim 9

-7. A speech recognition system comprising: -means for making an acoustic description of a given portion of speech to be recognized, as spoken by a given group of one or more speakers; -means for storing a plurality of individually trained acoustic word models, each of which is associated with a given word, and each of which is derived from acoustic data produced by having one or more speakers from said given group speak one or more utterances of its associated word; -means for storing a plurality of phonetic acoustic word models, each of which is associated with a given word, and none of which are derived from acoustic data produced by having any speakers from said given group speak its associated word; -recognition means for comparing both said individually trained and said phonetic acoustic word models against said acoustic description of a given portion of speech to be recognized and for selecting which one or more of said models best match said acoustic description.

Claim 10

-7. A speech recognition system comprising: -means for making an acoustic description of a given portion of speech to be recognized, as spoken by a given group of one or more speakers; -means for storing a plurality of individually trained acoustic word models, each of which is associated with a given word, and each of which is derived from acoustic data produced by having one or more speakers from said given group speak one or more utterances of its associated word; -means for storing a plurality of phonetic acoustic word models, each of which is associated with a given word, and none of which are derived from acoustic data produced by having any speakers from said given group speak its associated word; -recognition means for comparing both said individually trained and said phonetic acoustic word models against said acoustic description of a given portion of speech to be recognized and for selecting which one or more of said models best match said acoustic description.

Claim 11

-8. A speech recognition system comprising: -means for making an acoustic description of a given portion of speech to be recognized; -means for storing a first, acoustically selectable, set of machine responses, each of which is associated with a word; -means for storing an acoustic word model of the word associated with each of said acoustically selectable machine responses; means for storing a second, non-acoustically selectable, set of machine responses, each of which is associated with a word; recognition means for selecting which one or more of said acoustic models best match said acoustic description; recognition indicating means for indicating to a user the corresponding one or more acoustically selectable machine responses associated with those best matching models; -filtering means for selecting a subset of said non-acoustically selectable machine responses, said filtering means making said selection without performing a match between said acoustic description and any acoustic word models; -filtering indicating means for indicating to the user the non-­acoustically selectable machine responses selected by the filtering means; -means for enabling a user to select one of the indicated acoustically selectable or non-acoustically selectable machine responses as a desired machine response.

Claim 12

-8. A speech recognition system comprising: -means for making an acoustic description of a given portion of speech to be recognized; -means for storing a first, acoustically selectable, set of machine responses, each of which is associated with a word; -means for storing an acoustic word model of the word associated with each of said acoustically selectable machine responses; means for storing a second, non-acoustically selectable, set of machine responses, each of which is associated with a word; recognition means for selecting which one or more of said acoustic models best match said acoustic description; recognition indicating means for indicating to a user the corresponding one or more acoustically selectable machine responses associated with those best matching models; -filtering means for selecting a subset of said non-acoustically selectable machine responses, said filtering means making said selection without performing a match between said acoustic description and any acoustic word models; -filtering indicating means for indicating to the user the non-­acoustically selectable machine responses selected by the filtering means; -means for enabling a user to select one of the indicated acoustically selectable or non-acoustically selectable machine responses as a desired machine response.

Claim 15

A speech recognition system which includes the system for enabling a user to create word models described in Claim 9, said speech recognition system further including: means for representing a body of text comprised of one or more words and for representing a word insertion location relative to said text; - means for storing a word in association with each of said machine responses; -recognition means for recognising a spoken word by selecting a word which matches said spoken word; and -means for inserting a representation of either the word selected by said recognition means or the word associated with the indicated machine response selected by the user into said body of text at said word insertion location.

Claim 18

A continuous speech recognition system comprising: -means for making an acoustic description of a given portion of speech to be recognized; -means for storing acoustic models of a plurality of words; -recognition means for matching sequences of acoustic word models against said portion of speech and for selecting a plurality of the best matching word sequences, each representing a sequence of words whose sequence of corresponding acoustic models provides one of the best matches against said acoustic description; -means for indicating to a user each of said plurality of best matching word sequences; and -means for enabling the user to select one of the indicated best matching word sequences for use as an output, without requiring the user to enter each word in the selected sequence.

Claim 19

A continuous speech recognition system comprising: -means for making an acoustic description of a given portion of speech to be recognized; -means for storing acoustic models of a plurality of words; -recognition means for matching sequences of acoustic word models against said portion of speech and for selecting a plurality of the best matching word sequences, each representing a sequence of words whose sequence of corresponding acoustic models provides one of the best matches against said acoustic description; -means for indicating to a user each of said plurality of best matching word sequences; and -means for enabling the user to select one of the indicated best matching word sequences for use as an output, without requiring the user to enter each word in the selected sequence.

Claim 20

A continuous speech recognition system comprising: -means for making an acoustic description of a given portion of speech to be recognized; -means for storing acoustic models of a plurality of words; -recognition means for matching sequences of acoustic word models against said acoustic description and for selecting a best matching word sequence, representing a sequence of words whose sequence of corresponding acoustic models provide one of the best matches against said acoustic description; -means for indicating the words of said best matching word sequence to a user; -means for enabling the user to select an individual word from the indicated best matching sequence of words; -means for enabling the user to correct the indicated best matching word sequence by correcting the selected word; -means for using the indicated best matching word sequence, with the corrected selected word as an output.

Claim 21

A continuous speech recognition system comprising: -means for making an acoustic description of a given portion of speech to be recognized; -means for storing acoustic models of a plurality of words; -recognition means for matching sequences of acoustic word models against said acoustic description and for selecting a best matching word sequence, representing a sequence of words whose sequence of corresponding acoustic models provide one of the best matches against said acoustic description; -means for indicating the words of said best matching word sequence to a user; -means for enabling the user to select an individual word from the indicated best matching sequence of words; -means for enabling the user to correct the indicated best matching word sequence by correcting the selected word; -means for using the indicated best matching word sequence, with the corrected selected word as an output.

Claim 22

A speech recognition system comprising: -means for storing a list of machine responses, each of which has associated with it a spelling comprised of a sequence of characters and a pronunciation; -means for storing an acoustic model for the pronunciation associated with each of the machine responses; -means for making an acoustic description of a portion of speech to be recognized; -means for enabling a user to enter a string of one or more characters as filtering information; -means for enabling the user to edit said string of characters once it has been entered; -filtering means for responding to the entry and editing of said string by selecting a subset of machine responses associated with spellings which contain the string of one or more characters as entered and edited by said user; and -recognition means for making a filtered selection of which one or more of said acoustic models best match said acoustic description of said portion of speech to be recognized, including means for causing the selection by said recognition means to favor the selection of acoustic models whose associated machine responses are in said subset selected by said filtering means.

Claim 23

A speech recognition system comprising: -means for storing a list of machine responses, each of which has associated with it a spelling comprised of a sequence of characters and a pronunciation; -means for storing an acoustic model for the pronunciation associated with each of the machine responses; -means for making an acoustic description of a portion of speech to be recognized; -means for enabling a user to enter a string of one or more characters as filtering information; -means for enabling the user to edit said string of characters once it has been entered; -filtering means for responding to the entry and editing of said string by selecting a subset of machine responses associated with spellings which contain the string of one or more characters as entered and edited by said user; and -recognition means for making a filtered selection of which one or more of said acoustic models best match said acoustic description of said portion of speech to be recognized, including means for causing the selection by said recognition means to favor the selection of acoustic models whose associated machine responses are in said subset selected by said filtering means.

Claim 24

A speech recognition system designed to recognise a series of spoken words, said system comprising: language model means for indicating the probability that a given word to be recognised will be each of a plurality of vocabulary words based on statistical information on the frequency of that word's use; - means for making an acoustic description of the utterance of each of said series of spoken words to be recognised; - means for storing acoustic models of a plurality of vocabulary words; - recognition means for selecting which one or more of said vocabulary word's acoustic models best match a given acoustic description of a word to be recognised, based both on the closeness of the match between said acoustic models and said acoustic description and on the probability indications by said language model means; - means for using the selection by said recognition means of one or more vocabulary words as best matching said given acoustic description to update the statistical information on the frequency of said one or more vocabulary words in said language model means and for causing said recognition means to use a probability based on said updated statistical information in the recognition of which vocabulary words best match acoustic descriptions of subsequent words.

Claim 25

A speech recognition system designed to recognise a series of spoken words, said system comprising: language model means for indicating the probability that a given word to be recognised will be each of a plurality of vocabulary words based on statistical information on the frequency of that word's use; - means for making an acoustic description of the utterance of each of said series of spoken words to be recognised; - means for storing acoustic models of a plurality of vocabulary words; - recognition means for selecting which one or more of said vocabulary word's acoustic models best match a given acoustic description of a word to be recognised, based both on the closeness of the match between said acoustic models and said acoustic description and on the probability indications by said language model means; - means for using the selection by said recognition means of one or more vocabulary words as best matching said given acoustic description to update the statistical information on the frequency of said one or more vocabulary words in said language model means and for causing said recognition means to use a probability based on said updated statistical information in the recognition of which vocabulary words best match acoustic descriptions of subsequent words.