US5503560A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Language training
Assignee: British Telecommunications PLC
•
Filing: 1989-07-25
•
Granted: 1996-04-02
•
Expires: 2009-07-25
Abstract
Abstract A speech synthesizer (3) produces prompts in the voice of a native speaker of a language to be learned to which the student replies or imitates. A phrase recognizer (1) employs keyword recognition to generate from the student's prompted response an original speech template spoken by the student. Thereafter, interactive dialogue takes place. The student's progress in that dialogue is monitored by measuring the deviation of the student's current speech from his original speech template. When this deviation is sufficiently large so that the recognizer (1) no longer recognizes what the student is saying, the system retrains and updates the student's speech template. In another embodiment, the system includes a display which shows the native speaker's mouth shape while the words to be imitated are spoken by the speech synthesizer (3). Also provided are a video pick-up and analyzer for analyzing the shapes of the student's mouth to give the student visual feedback.
Inventors
- 👤 Frederick W. M. Stentiford
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-30 17:18
CPC Classifications
G; G09; G09B; G09B5/00; G09B5/04; G09B19/00; G09B19/06; H; H04; H04B; H04B10/00; H04B10/25; H04B10/2589
Claims Analysis
Total Claims: 25
Claim 1
Independent Claim
98 words
Apparatus for pronunciation training comprising: a speech generator for generating utterances to prompt a trainee; a controller for generating speech templates from utterances by the trainee in response to prompts from the speech generator of words from a predetermined selected set of words; a speech recognizer for recognizing, in a trainee's current utterances, words from the predetermined selected set of words by comparing each of the trainee's current utterances with said speech templates; and an output device indicating improvement in the trainee's pronunciation based on increases in a deviation of the trainee's current utterances from the speech templates.
Claim 2
Dependent Claim
98 words
Apparatus for pronunciation training comprising: a speech generator for generating utterances to prompt a trainee; a controller for generating speech templates from utterances by the trainee in response to prompts from the speech generator of words from a predetermined selected set of words; a speech recognizer for recognizing, in a trainee's current utterances, words from the predetermined selected set of words by comparing each of the trainee's current utterances with said speech templates; and an output device indicating improvement in the trainee's pronunciation based on increases in a deviation of the trainee's current utterances from the speech templates.
Claim 3
Dependent Claim
25 words
Apparatus according to claim 1, wherein the controller updates the speech templates from the trainee's current utterances when the said deviation exceeds a predetermined value.
Claim 4
Dependent Claim
46 words
Apparatus according to claim 1, wherein the speech recognizer recognizes in the trainee's response to the prompt one or more words from the set of words, and wherein the speech generator generates an utterance depending on the one or more words recognized by the speech recognizer.
Claim 5
Dependent Claim
69 words
Apparatus for pronunciation training according to claim 3, further comprising: phrase recognition means for identifying phrases by a combination and order of words from the predetermined selected set, wherein when the trainee is prompted by an utterance generated by the speech generator to utter a phrase, the phrase recognition means recognizes the phrase and selects the utterance generated by the speech generator to be a reply to the phrase.
Claim 6
Dependent Claim
34 words
A pronunciation training apparatus according to claim 1, further comprising: a video generator for generating corresponding video images of a mouth to prompt a trainee to imitate a correct pronunciation of the generated utterances.
Claim 7
Dependent Claim
34 words
Apparatus according to claim 5, further comprising: a video analyzer for analyzing mouth movements of the trainee and displaying corresponding synthesized and analyzed mouth movements generated by the video generator and by the trainee.
Claim 8
Dependent Claim
26 words
Language training apparatus according to claim 1, wherein the speech generator generates utterances in a language in the accent of a native speaker of that language.
Claim 9
Independent Claim
67 words
A method of pronunciation training, comprising; prompting a trainee to pronounce a series of words; generating trainee speech templates from the trainee's pronunciations; prompting a trainee to speak an utterance; analyzing the utterance using the trainee speech templates; and assessing improvements in pronunciation by measuring a difference between the utterance and a corresponding trainee speech template, wherein an increase in difference corresponds to an improvement in pronunciation.
Claim 10
Independent Claim
67 words
A method of pronunciation training, comprising; prompting a trainee to pronounce a series of words; generating trainee speech templates from the trainee's pronunciations; prompting a trainee to speak an utterance; analyzing the utterance using the trainee speech templates; and assessing improvements in pronunciation by measuring a difference between the utterance and a corresponding trainee speech template, wherein an increase in difference corresponds to an improvement in pronunciation.
Claim 11
Dependent Claim
22 words
A method according to claim 8 further comprising the step of: updating the speech template when the difference exceeds a predetermined value.
Claim 12
Independent Claim
74 words
A system for training a speaker's pronunciation, comprising: a speech synthesizer for generating synthesized speech to prompt the speaker to generate speech samples; control circuiting for determining speech pronunciation characteristics from the speaker's initial speech samples generated in response to speech synthesizer prompts; processing subsequent speech samples of the speaker to determine a deviation from the determined speech pronunciation characteristics and indicating progress in the speaker's pronunciation based on an increase in the deviation.
Claim 13
Independent Claim
74 words
A system for training a speaker's pronunciation, comprising: a speech synthesizer for generating synthesized speech to prompt the speaker to generate speech samples; control circuiting for determining speech pronunciation characteristics from the speaker's initial speech samples generated in response to speech synthesizer prompts; processing subsequent speech samples of the speaker to determine a deviation from the determined speech pronunciation characteristics and indicating progress in the speaker's pronunciation based on an increase in the deviation.
Claim 14
Dependent Claim
22 words
The system according to claim 10, further comprising: means for modifying the derived pronunciation characteristics when the deviation exceeds a predetermined threshold.
Claim 15
Dependent Claim
34 words
The system according to claim 10, wherein said speech synthesizer prompts speech from the speaker in a foreign language and the means for receiving receives speech samples from the speaker in the foreign language.
Claim 16
Dependent Claim
16 words
The system according to claim 10, wherein the speech pronunciation characteristics include key words and phrases.
Claim 17
Dependent Claim
45 words
The system according to claim 10, wherein the speech synthesizer prompts the speaker using words in a native language of the speaker and the means for processing processes subsequent responsive speech samples received in a foreign language corresponding to translations of the native language words.
Claim 18
Dependent Claim
35 words
The system according to claim 10, wherein the speech synthesizer prompts the speaker using words in a foreign language and the control circuitry processes subsequent speech samples received from the speaker in a foreign language.
Claim 19
Dependent Claim
25 words
The system according to claim 10, wherein the speech synthesizer generates words and phrases using the accent of a native speaker of a foreign language.
Claim 20
Dependent Claim
37 words
The system according to claim 10, further comprising: a video generator for generating video images of a human mouth configured to correspond with proper pronunciation of the synthesized speech samples to assist the speaker in a response.
Claim 21
Independent Claim
110 words
A method of pronunciation training comprising: prompting a trainee using words generated by a speech generator in a secondary language different than a primary language of the trainee as pronounced by a native speaker of said secondary language, the trainee pronouncing speech in the secondary language in response to said prompting; deriving trainee speech templates from the speech in the secondary language pronounced by said trainee; prompting dialog responses from said trainee in said secondary language via said speech generator; comparing said dialog responses to corresponding ones of said trainee speech templates; and indicating progress in pronunciation as differences between said dialog responses and said corresponding trainee speech templates increase.
Claim 22
Independent Claim
110 words
A method of pronunciation training comprising: prompting a trainee using words generated by a speech generator in a secondary language different than a primary language of the trainee as pronounced by a native speaker of said secondary language, the trainee pronouncing speech in the secondary language in response to said prompting; deriving trainee speech templates from the speech in the secondary language pronounced by said trainee; prompting dialog responses from said trainee in said secondary language via said speech generator; comparing said dialog responses to corresponding ones of said trainee speech templates; and indicating progress in pronunciation as differences between said dialog responses and said corresponding trainee speech templates increase.
Claim 23
Dependent Claim
26 words
The method according to claim 18, further comprising: updating said trainee templates when a difference between said dialog responses and said trainee templates exceeds a threshold.
Claim 24
Dependent Claim
29 words
The method according to claim 18, wherein said dialog generating step includes: generating a video image of said native speaker's mouth shape synchronized with said generated secondary dialog speech.
Claim 25
Dependent Claim
20 words
The method according to claim 18, further comprising: analyzing shapes of the mouth of said trainee during said dialog responses.
References Cited (Backward Citations)
US3703602A
Cited
US3946504A
Cited
EP0094502A1
Cited
US4507750A
Cited
US4615680A
Cited
GB2167224A
Cited
EP0225729A1
Cited
EP0262938A1
Cited
US4884972A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1999-04-02 | 1999-10-02 | 2000-04-02 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2003-04-02 | 2003-10-02 | 2004-04-02 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2007-04-02 | 2007-10-02 | 2008-04-02 | $7700.00 | Window Closed / Lapsed |