Context sharing of similarities in context dependent word models
First Claim
1. A method for automatic speech recognition comprising the steps of:
- building a model for a vocabulary of sounds wherein at least two sounds share a common head or a common tail;
wherein said vocabulary of sounds includes subwords;
receiving an utterance containing at least one word;
processing the utterance into cepstral coefficients; and
recognizing at least one word in the utterance using said model.
7 Assignments
0 Petitions
Accused Products
Abstract
A natural number recognition method and system that uses minimum classification error trained inter-word context dependent models of the head-body-tail type over a specific vocabulary. One part of the method and system allows recognition of spoken monetary amounts in financial transactions. A second part of the method and system allows recognition of numbers such as credit card or U.S. telephone numbers. A third part of the method and system allows recognition of natural language expressions of time, such as time of day, day of the week and date of the month for applications such as scheduling or schedule inquires. Even though limited natural language expressions are allowed, context sharing between similar sounds in the vocabulary within a head-body-tail model keeps storage and processing time requirements to manageable levels.
-
Citations
22 Claims
-
1. A method for automatic speech recognition comprising the steps of:
-
building a model for a vocabulary of sounds wherein at least two sounds share a common head or a common tail;
wherein said vocabulary of sounds includes subwords;
receiving an utterance containing at least one word;
processing the utterance into cepstral coefficients; and
recognizing at least one word in the utterance using said model. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12)
-
-
13. A method for automatic speech recognition comprising the steps of:
-
receiving an utterance containing at least one word of a vocabulary of words;
processing the utterance into cepstral coefficients;
separating the utterance into a plurality of words;
separating at least one of said plurality of words into a head portion, a body portion and a tail portion;
recognizing at least one word from the vocabulary using said head portion, said body portion and said tail portion. - View Dependent Claims (14, 15, 16)
-
-
17. A method for automatic speech recognition comprising the steps of:
-
receiving an utterance containing at least one digit word and at least one non-digit word;
processing the utterance into cepstral coefficients;
separating the utterance into a plurality of words;
separating at least one of said plurality of words into a head portion, a body portion and a tail portion;
recognizing said at least one word using a vocabulary for numbers, dates and times of day. - View Dependent Claims (18, 19, 20, 21, 22)
a shared context of numbers ending in letters ‘
teen’a shared context of numbers ending in letters ‘
ty’
, anda shared context of numbers beginning in letters ‘
seven’
.
-
-
19. The method of claim 17, wherein for scheduling said plurality of contexts includes:
-
a shared context of months ending in letters ‘
ber’a shared context of days of the week ending in the letters ‘
day’a shared context of numbers ending in letters ‘
teen’a shared context of numbers ending in letters ‘
ty’
, anda shared context of numbers beginning in letters ‘
seven’
.
-
-
20. The method of claim 17, wherein said vocabulary words comprise spoken numbers.
-
21. The method of claim 17, wherein said vocabulary words comprise spoken dates.
-
22. The method of claim 17, wherein said vocabulary words comprise spoken times.
Specification
-
- Resources
-
Current AssigneeWSOU Investments, LLC (WSOU Holdings, LLC)
-
Original AssigneeLucent Technologies, Inc. (Nokia Corporation)
-
InventorsJacob, John, Gandhi, Malan Bhatki
-
Primary Examiner(s)Dorvil, Richemond
-
Assistant Examiner(s)NOLAN, DANIEL A
-
Application NumberUS09/184,620Time in Patent Office1,037 DaysField of Search704/251, 704/252, 704/253-256, 704/9, 704/10US Class Current704/256CPC Class CodesG10L 15/142 Hidden Markov Models [HMMs]G10L 15/18 using natural language mode...
