← Back to Patent Explorer
US5806033A Expired / Lapsed US ⏱️ 0.0 years remaining

Syllable duration and pitch variation to determine accents and stresses for speech recognition

Assignee: Intellectual Ventures I LLC
•
Filing: 1996-06-17
•
Granted: 1998-09-08
•
Expires: 2016-06-17
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract A method and device for speech-to-text conversion. From a given speech, the fundamental tone is extracted. A model of the speech is further created from the speech. In the model, a duration reproduction in words and sentences is obtained. The duration reproduction is compared with a segment duration in the speech. Information which decides which type of accent exists is obtained from the comparison from which a text with sentence accent information is produced. Further, in creating the model lexical information, which may indicate accents in words, can be utilized. Syntax analysis of the model can further by utilized to model a fundamental tone in a sentence. Such operations allow the re-recognition utilizing the accent and stress after a decision as to the type of accent which exists.

Inventors

  • 👤 Bertil Lyberg

Bibliographic Details

Application Number —
Jurisdiction US
Data Source google_patents
Last Synchronized 2026-11-06 05:34
CPC Classifications
G; G10; G10L; G10L15/00; G10L15/08; G10L15/18; G10L15/1807

Claims Analysis

Total Claims: 26
Claim 1
Independent Claim 69 words
Method for speech-to-text-conversion by extracting a fundamental tone from speech, and creating a model of the speech from the speech, comprising the steps of: obtaining from the model a duration reproduction in words and sentences, comparing the duration reproduction with a segment duration in the speech, deciding from the comparison which type of accent exists, and producing a text with sentence accent information from results of the deciding step.
Claim 2
Dependent Claim 69 words
Method for speech-to-text-conversion by extracting a fundamental tone from speech, and creating a model of the speech from the speech, comprising the steps of: obtaining from the model a duration reproduction in words and sentences, comparing the duration reproduction with a segment duration in the speech, deciding from the comparison which type of accent exists, and producing a text with sentence accent information from results of the deciding step.
Claim 3
Dependent Claim 21 words
Method according to claim 1, wherein the step of deciding comprises discerning a first accent, a second accent, and sentence accents.
Claim 4
Dependent Claim 16 words
Method according to claim 1, wherein the step of deciding comprises discerning stressed and unstressed syllables.
Claim 5
Dependent Claim 22 words
Method according to claim 1, wherein the step of creating a model comprises utilizing, at the creation of the model lexical information.
Claim 6
Dependent Claim 15 words
Method according to claim 4, wherein the lexical information indicates alternative accents in the words.
Claim 7
Dependent Claim 24 words
Method according to claim 4, wherein the lexical information indicates alternative durations for different segments in the words which are obtained from the model.
Claim 8
Dependent Claim 22 words
Method according to claim 1, wherein the step of obtaining comprises modeling a model of the fundamental tone in words and sentences.
Claim 9
Dependent Claim 28 words
Method according to claim 7, wherein the step of comparing comprises comparing an extracted fundamental tone with the modelled fundamental tone to obtain an indication of possible accents.
Claim 10
Dependent Claim 38 words
Method according to claim 8, wherein the step of comparing compares the possible accents at the fundamental tones and the duration, and wherein the step of deciding decides which type of accent or stressed or unstressed syllable exists.
Claim 11
Dependent Claim 32 words
Method according to claim 9, further comprising the step of using the decision for correction of the model to produce text from which one gets the corresponding speech meaning of the speech.
Claim 12
Dependent Claim 21 words
Method according to claim 7, wherein syntax analysis of the model is utilized to model the fundamental tone in the sentence.
Claim 13
Dependent Claim 18 words
Method according to claim 11, wherein the syntax analysis of the model is utilized to model the sentences.
Claim 14
Dependent Claim 89 words
Device for speech-to-text conversion, from a fundamental tone extracted from speech in a speech recognition equipment, and from a model of the speech created in the speech recognition equipment, the device comprising: means for creating a duration reproduction in words and sentences from the model, a comparator for comparing a duration reproduction with a segment duration in the speech, means for deciding an accent type based on a result of the comparator, and means for producing a text with sentence accent information from results of the means for deciding.
Claim 15
Dependent Claim 89 words
Device for speech-to-text conversion, from a fundamental tone extracted from speech in a speech recognition equipment, and from a model of the speech created in the speech recognition equipment, the device comprising: means for creating a duration reproduction in words and sentences from the model, a comparator for comparing a duration reproduction with a segment duration in the speech, means for deciding an accent type based on a result of the comparator, and means for producing a text with sentence accent information from results of the means for deciding.
Claim 16
Dependent Claim 23 words
Device for speech-to-text conversion according to claim 13, wherein the comparator comprises a comparator for discerning accent 1, accent 2, and sentence accents.
Claim 17
Dependent Claim 20 words
Device for speech-to-text conversion according to claim 13, wherein the comparator comprises a comparator for discerning stressed and unstressed syllables.
Claim 18
Dependent Claim 17 words
Device for speech-to-text conversion according to claim 13, wherein lexical information is utilized to create the model.
Claim 19
Dependent Claim 22 words
Device for speech-to-text conversion according to claim 16, further comprising means to obtain alternative accents in the words from the lexical information.
Claim 20
Dependent Claim 25 words
Device for speech-to-text conversion according to claim 16, further comprising means to obtain alternative durations for different segments in recognized words from the lexical information.
Claim 21
Dependent Claim 24 words
Device for speech-to-text conversion according to claim 13, further comprising a modeller for producing a model of the fundamental tone in words and sentences.
Claim 22
Dependent Claim 35 words
Device for speech-to-text conversion according to claim 19, wherein the comparator further comprises a comparator for comparing the extracted fundamental tone with the modelled fundamental tone to obtain an indication of possible locations of accents.
Claim 23
Dependent Claim 41 words
Device for speech-to-text conversion according to claim 13, wherein the comparator comprises a comparator for comparing possible accents of the fundamental tone; and wherein the means for deciding comprises a means for deciding which type of accent or stressed/unstressed syllable exists.
Claim 24
Dependent Claim 23 words
Device for speech-to-text conversion according to claim 13, wherein during modelling of the fundamental tone, a syntax analysis of the model is utilized.
Claim 25
Dependent Claim 22 words
Device for speech-to-text conversion according to claim 13, wherein during modelling of the sentences, the syntax analysis of the model is utilized.
Claim 26
Dependent Claim 29 words
Device for speech-to-text conversion according to claim 13, further comprising a means for using a result from the means for deciding to correct the model and to produce text.

References Cited (Backward Citations)

US4441202A Cited
EP0527650A2 Cited
US5546500A Cited

USPTO Annuity & Maintenance Windows

Annuity Stage Window Opens Payment Deadline Surcharge Period Until Estimated Fee Status
3.5 Years (Year 4) 2001-09-07 2002-03-09 2002-09-08 $2000.00 Window Closed / Lapsed
7.5 Years (Year 8) 2005-09-07 2006-03-09 2006-09-08 $3760.00 Window Closed / Lapsed
11.5 Years (Year 12) 2009-09-07 2010-03-09 2010-09-08 $7700.00 Window Closed / Lapsed

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag