US6185530B1
Expired / Lapsed
US
⏱️ 0.0 years remaining
Apparatus and methods for identifying potential acoustic confusibility among words in a speech recognition system
Assignee: Nuance Communications Inc
•
Filing: 1998-08-14
•
Granted: 2001-02-06
•
Expires: 2018-08-14
Abstract
Abstract A method of determining potential acoustic confusion between at least one new word and at least a portion of existing words of a vocabulary of a speech recognition engine comprises the steps of: a user inputting the at least one new word; computing respective measures between the at least one new word and the at least a portion of existing vocabulary words, the respective measures indicative of acoustic similarity between the at least one word and the at least a portion of existing words; if at least one measure is within a threshold range, indicating results associated with the at least one measure and prompting the user to input an alternative word or additional information pertaining to the at least one new word; and if no measure is within the threshold range, adding the at least one new word to the vocabulary.
Inventors
- 👤 Abraham Ittycheriah
- 👤 Stephane H. Maes
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 13:27
CPC Classifications
G; G10; G10L; G10L15/00; G10L15/08; G10L15/10; Y; Y10; Y10S; Y10S707/00; Y10S707/99931; Y10S707/99933; Y10S707/99935; Y10S707/99936
Claims Analysis
Total Claims: 33
Claim 1
Independent Claim
226 words
A method for adding words to a vocabulary of a speech recognition engine, the method comprising the steps of: (a) inputting at least one new word; (b) computing an acoustic similarity measure between the at least one new word and each word of at least a portion of existing vocabulary words, wherein the step of computing an acoustic similarity measure comprises the steps of: computing a leaf sequence for the at least one new word; comparing the leaf sequence for the at least one new word to respective leaf sequences associated with the at least a portion of the existing words; and generating respective distance measures in response to the comparisons, the respective distance measures indicative of acoustic distances between the compared leaf sequences, wherein the respective distance measures are calculated via a Kuhlback-Liebler distance metric; (c) if at least one of the computed acoustic similarity measures is within a threshold range, identifying the at least one new word as being potentially acoustically confusing with an existing vocabulary word, indicating results associated with the at least one computed acoustic similarity measure and prompting a user to input an alternative word or additional information pertaining to the at least one new word; and (d) if none of the computed acoustic similarity measures are within the threshold range, adding the at least one new word to the vocabulary.
Claim 2
Independent Claim
226 words
A method for adding words to a vocabulary of a speech recognition engine, the method comprising the steps of: (a) inputting at least one new word; (b) computing an acoustic similarity measure between the at least one new word and each word of at least a portion of existing vocabulary words, wherein the step of computing an acoustic similarity measure comprises the steps of: computing a leaf sequence for the at least one new word; comparing the leaf sequence for the at least one new word to respective leaf sequences associated with the at least a portion of the existing words; and generating respective distance measures in response to the comparisons, the respective distance measures indicative of acoustic distances between the compared leaf sequences, wherein the respective distance measures are calculated via a Kuhlback-Liebler distance metric; (c) if at least one of the computed acoustic similarity measures is within a threshold range, identifying the at least one new word as being potentially acoustically confusing with an existing vocabulary word, indicating results associated with the at least one computed acoustic similarity measure and prompting a user to input an alternative word or additional information pertaining to the at least one new word; and (d) if none of the computed acoustic similarity measures are within the threshold range, adding the at least one new word to the vocabulary.
Claim 3
Dependent Claim
31 words
The method of claim 1 , further comprising the step of checking whether the at least one new word already exists within the vocabulary, prior to computing the acoustic similarity measures.
Claim 4
Dependent Claim
47 words
The method of claim 2 , further comprising the step of indicating previously accumulated information relating to the at least one new word, if the at least one new word already exists within the vocabulary, the accumulated information comprising previously catalogued decoding problems associated with the word.
Claim 5
Dependent Claim
27 words
The method of claim 1 , wherein the leaf sequence comparison step further comprises performing a best match alignment process between leaf sequences of unequal phonetic length.
Claim 6
Dependent Claim
21 words
The method of claim 1 , wherein the additional information pertaining to the at least one new word includes contextual information.
Claim 7
Dependent Claim
55 words
The method of claim 1 , wherein the additional information pertaining to the at least one new word includes an instruction by the user to temporarily exclude the existing word associated with a measure within the threshold range from the vocabulary when the at least one new word is uttered in a real-time decoding session.
Claim 8
Dependent Claim
33 words
The method of claim 1 , wherein the indicating step comprises displaying the results to the user, the results comprising the at least one acoustic similarity measure and the corresponding existing vocabulary word.
Claim 9
Dependent Claim
20 words
The method of claim 1 , wherein the indicating step comprises speech synthesizing the results for playback to the user.
Claim 10
Dependent Claim
35 words
The method of claim 1 , wherein the indicating step further comprises a step of prompting the user to request an additional search to provide increased identification of potentially acoustically confusing words in the vocabulary.
Claim 11
Dependent Claim
23 words
The method of claim 9 , wherein the additional search includes increasing a beamwidth associated with a Viterbi algorithm performed during the search.
Claim 12
Dependent Claim
62 words
The method of claim 1 , further comprising, between steps (b) and (c), the step of performing an additional search to provide increased identification of potentially acoustically confusing words in the vocabulary, if at least one of the acoustic similarity measures is within a threshold range, the additional search including increasing a bandwidth associated with a Viterbi algorithm performed during the search.
Claim 13
Dependent Claim
24 words
The method of claim 1 , wherein the inputting step comprises providing at least one baseform associated with the at least one new word.
Claim 14
Dependent Claim
18 words
The method of claim 12 , wherein the at least one baseform is provided via a speech synthesizer.
Claim 15
Dependent Claim
196 words
Computer-based apparatus for adding words to a vocabulary of a speech recognition engine, the apparatus comprising: an input device for inputting at least one new word; a processor, operatively coupled to the input device, for computing an acoustic similarity measure between the at least one new word and each word of at least a portion of existing vocabulary words; and an output device operatively coupled to the processor, wherein the output device comprises a text-to-speech system; wherein if at least one of the computed acoustic similarity measures is within a threshold range, the processor generates output results associated with the at least one computed acoustic similarity measure to identify the at least one new word as being potentially acoustically confusing with an existing vocabulary word, wherein the processor causes speech synthesis of the results for playback to the user via the text-to-speech system and causes a prompting of the user to input an alternative word or additional information pertaining to the at least one new word; and wherein if none of the computed acoustic similarity measures are within the threshold range, the processor causes the at least one new word to be added to the vocabulary.
Claim 16
Dependent Claim
196 words
Computer-based apparatus for adding words to a vocabulary of a speech recognition engine, the apparatus comprising: an input device for inputting at least one new word; a processor, operatively coupled to the input device, for computing an acoustic similarity measure between the at least one new word and each word of at least a portion of existing vocabulary words; and an output device operatively coupled to the processor, wherein the output device comprises a text-to-speech system; wherein if at least one of the computed acoustic similarity measures is within a threshold range, the processor generates output results associated with the at least one computed acoustic similarity measure to identify the at least one new word as being potentially acoustically confusing with an existing vocabulary word, wherein the processor causes speech synthesis of the results for playback to the user via the text-to-speech system and causes a prompting of the user to input an alternative word or additional information pertaining to the at least one new word; and wherein if none of the computed acoustic similarity measures are within the threshold range, the processor causes the at least one new word to be added to the vocabulary.
Claim 17
Dependent Claim
30 words
The apparatus of claim 14 , wherein the processor further checks whether the at least one new word already exists within the vocabulary, prior to computing the acoustic similarity measures.
Claim 18
Dependent Claim
49 words
The apparatus of claim 15 , wherein the processor further causes an output of previously accumulated information relating to the at least one new word, if the at least one new word already exists within the vocabulary, the accumulated information comprising previously catalogued decoding problems associated with the word.
Claim 19
Dependent Claim
76 words
The apparatus of claim 14 , wherein the processor further performs the steps of: (a) computing a leaf sequence for the at least one new word; (b) comparing the leaf sequence for the at least one new word to respective leaf sequences associated with the at least a portion of existing words; and (c) generating respective distance measures in response to the comparisons, the respective distance measures indicative of acoustic distances between the compared leaf sequences.
Claim 20
Dependent Claim
23 words
The apparatus of claim 17 , wherein the processor further performs a best match alignment process between leaf sequences of unequal phonetic length.
Claim 21
Dependent Claim
19 words
The apparatus of claim 17 , wherein the processor calculates the respective distance measures via a Kuhlback-Liebler distance metric.
Claim 22
Dependent Claim
21 words
The apparatus of claim 14 , wherein the additional information pertaining to the at least one new word includes contextual information.
Claim 23
Dependent Claim
55 words
The apparatus of claim 14 , wherein the additional information pertaining to the at least one new word includes an instruction by the user to temporarily exclude the existing word associated with a measure within the threshold range from the vocabulary when the at least one new word is uttered in a real-time decoding session.
Claim 24
Dependent Claim
46 words
The apparatus of claim 14 , wherein the output device is a display and further wherein the processor causes display of the results to the user on the display, the results comprising the at least one computed acoustic similarity measure and the corresponding existing vocabulary word.
Claim 25
Dependent Claim
31 words
The apparatus of claim 14 , wherein the processor causes prompting of the user to request an additional search to provide increased identification of potentially acoustically confusing words in the vocabulary.
Claim 26
Dependent Claim
23 words
The apparatus of claim 23 , wherein the additional search includes increasing a beamwidth associated with a Viterbi algorithm performed during the search.
Claim 27
Dependent Claim
63 words
The apparatus of claim 14 , wherein the processor, prior to generating the output results to the user, performs an additional search to provide increased identification of potentially acoustically confusing words in the vocabulary, if at least one computed acoustic similarity measure is within a threshold range, the additional search including increasing a beamwidth associated with a Viterbi algorithm performed during the search.
Claim 28
Dependent Claim
26 words
The apparatus of claim 14 , wherein the input device comprises means for inputting at least one baseform associated with the at least one new word.
Claim 29
Dependent Claim
14 words
The apparatus of claim 26 , wherein the inputting means is a speech synthesizer.
Claim 30
Dependent Claim
177 words
A program storage device, readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for adding words to a vocabulary of a speech recognition engine, the method comprising the steps of: (a) receiving as input at least one new word; (b) computing an acoustic similarity measure between the at least one new word and each word of at least a portion of existing vocabulary words; (c) if at least one of the computed acoustic similarity measures is within a threshold range, identifying the at least one new word as being potentially acoustically confusing with an existing vocabulary word, indicating results associated with the at least one computed acoustic similarity measure, wherein the indicating step comprises speech synthesizing the results for playback to the user, and prompting a user to input an alternative word or additional information pertaining to the at least one new word; and (d) if none of the computed acoustic similarity measures are within the threshold range, adding the at least one new word to the vocabulary.
Claim 31
Dependent Claim
177 words
A program storage device, readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for adding words to a vocabulary of a speech recognition engine, the method comprising the steps of: (a) receiving as input at least one new word; (b) computing an acoustic similarity measure between the at least one new word and each word of at least a portion of existing vocabulary words; (c) if at least one of the computed acoustic similarity measures is within a threshold range, identifying the at least one new word as being potentially acoustically confusing with an existing vocabulary word, indicating results associated with the at least one computed acoustic similarity measure, wherein the indicating step comprises speech synthesizing the results for playback to the user, and prompting a user to input an alternative word or additional information pertaining to the at least one new word; and (d) if none of the computed acoustic similarity measures are within the threshold range, adding the at least one new word to the vocabulary.
Claim 32
Independent Claim
176 words
A method for adding words to a vocabulary of a speech recognition engine, the method comprising the steps of: (a) receiving as input at least one new word, wherein the step of receiving as input comprises providing at least one baseform associated with the at least one new word, wherein the at least one baseform is provided via a speech synthesizer; (b) computing an acoustic similarity measure between the at least one new word and each word of at least a portion of existing vocabulary words; (c) if at least one of the computed acoustic similarity measures is within a threshold range, identifying the at least one new word as being potentially acoustically confusing with an existing vocabulary word, indicating results associated with the at least one computed acoustic similarity measure and prompting a user to input an alternative word or additional information pertaining to the at least one new word; and (d) if none of the computed acoustic similarity measures are within the threshold range, adding the at least one new word to the vocabulary.
Claim 33
Independent Claim
176 words
A method for adding words to a vocabulary of a speech recognition engine, the method comprising the steps of: (a) receiving as input at least one new word, wherein the step of receiving as input comprises providing at least one baseform associated with the at least one new word, wherein the at least one baseform is provided via a speech synthesizer; (b) computing an acoustic similarity measure between the at least one new word and each word of at least a portion of existing vocabulary words; (c) if at least one of the computed acoustic similarity measures is within a threshold range, identifying the at least one new word as being potentially acoustically confusing with an existing vocabulary word, indicating results associated with the at least one computed acoustic similarity measure and prompting a user to input an alternative word or additional information pertaining to the at least one new word; and (d) if none of the computed acoustic similarity measures are within the threshold range, adding the at least one new word to the vocabulary.
References Cited (Backward Citations)
US5218668A
Cited
US4918732A
Cited
US4829576A
Cited
US5033087A
Cited
US5621857A
Cited
US5349645A
Cited
US5675704A
Cited
US5850627A
Cited
US5625748A
Cited
US5864810A
Cited
US5715367A
Cited
US5752001A
Cited
US5680511A
Cited
US6005549A
Cited
US5852801A
Cited
US5778344A
Cited
US6023673A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2004-02-06 | 2004-08-07 | 2005-02-06 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2008-02-06 | 2008-08-07 | 2009-02-06 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2012-02-06 | 2012-08-07 | 2013-02-06 | $7700.00 | Window Closed / Lapsed |