US5101434A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Voice recognition using segmented time encoded speech
Assignee: HYDRALOGICA IP Ltd
•
Filing: 1988-09-01
•
Granted: 1992-03-31
•
Expires: 2008-09-01
Abstract
Abstract PCT No. PCT/GB88/00718 Sec. 371 Date Jan. 4, 1990 Sec. 102(e) Date Jan. 4, 1990 PCT Filed Sep. 1, 1988 PCT Pub. No. WO89/02146 PCT Pub. Date Mar. 9, 1989.A voice recognition system for time encoded speech (TES). The TES coded stream is divided into time frames. Each time frame is fed to a number of seeker circuits, each seeker circuit being optimized to detect an acoustic event of the voice signals, such as voiced or unvoiced events, amplitude or silence (ambient noise). Segmentation boundaries in the TES symbol stream are then determined from the acoustic events sensed as present or absent. These events are compared with stored events to determine the voice signals. By using this parallel comparison technique, comparison is made using distinct boundaries determined from the voice signal content and not by arbitrary boundaries which may contain acoustic events from, for example, more than one word in the voice signals. The implementation of the TES recognizer in low cost complimentary metal-oxide semiconductor (CMOS) technology makes this parallel comparison technique feasible at a reasonable cost.
Inventors
- 👤 Reginald A. King
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-30 09:32
CPC Classifications
G; G10; G10L; G10L15/00
Claims Analysis
Total Claims: 22
Claim 1
Independent Claim
146 words
A method of recognizing voice signals encoded as time encoded speech, the method comprising the steps of: segmenting a time encoded speech symbol stream into a number of time frames and applying each time frame to a plurality of seeker circuits, each seeker circuit being optimized to detect an acoustic event of the voice signals to be recognized, examining the time encoded speech symbols by means of the seeker circuits in the time frames to determine the presence or absence of any acoustic event to which any seeker circuit is optimized, determining segmentation boundaries in the time encoded speech symbol stream in dependence upon the acoustic events determined as present or absent, and comparing the time encoded speech symbol stream within the segmentation boundaries with archetypes of words or utterances stored as time encoded speech to generate an indication of the nature of the voice signal.
Claim 2
Independent Claim
146 words
A method of recognizing voice signals encoded as time encoded speech, the method comprising the steps of: segmenting a time encoded speech symbol stream into a number of time frames and applying each time frame to a plurality of seeker circuits, each seeker circuit being optimized to detect an acoustic event of the voice signals to be recognized, examining the time encoded speech symbols by means of the seeker circuits in the time frames to determine the presence or absence of any acoustic event to which any seeker circuit is optimized, determining segmentation boundaries in the time encoded speech symbol stream in dependence upon the acoustic events determined as present or absent, and comparing the time encoded speech symbol stream within the segmentation boundaries with archetypes of words or utterances stored as time encoded speech to generate an indication of the nature of the voice signal.
Claim 3
Dependent Claim
21 words
A method according to claim 1 wherein the segmentation boundaries comprise adjacent segmentation boundaries in the time encoded speech symbol stream.
Claim 4
Dependent Claim
25 words
A method according to claim 1 wherein an acoustic event of the voice signals to which at least one seeker circuit is optimized comprises silence.
Claim 5
Dependent Claim
25 words
A method according to claim 1 wherein an acoustic event of the voice signals to which at least one seeker circuit is optimized comprises amplitude.
Claim 6
Dependent Claim
26 words
A method according to claim 1 wherein an acoustic event of the voice signals to which at least one seeker circuit is optimized comprises voice events.
Claim 7
Dependent Claim
26 words
A method according to claim 1 wherein an acoustic event of the voice signals to which at least one seeker circuit is optimized comprises unvoiced events.
Claim 8
Dependent Claim
29 words
A method according to claim 6 wherein the voice signals are differentiated before being processed by the seeker circuits to emphasize any high frequency component of the voice signals.
Claim 9
Dependent Claim
29 words
A method according to claim 6 wherein the voice signals are integrated before being processed by the seeker circuits to emphasize any low frequency component of the voice signals.
Claim 10
Dependent Claim
75 words
A method according to claim 8 and further comprising the step of arranging each seeker circuit to operate in accordance with a classification algorithm whereby each seeker circuit searches for regions of consistent contiguous acoustic events in the time encoded symbol stream such that the segmentation boundaries are placed between such regions to define segments in the time encoded symbol stream and wherein the length of any segment so determined exceeds a predetermined minimum length.
Claim 11
Dependent Claim
56 words
A method according to claim 9 and further comprising the steps of providing a matrix to be compiled for any segment being determined in the time encoded symbol stream, and, generating acoustic events for each matrix to identify each matrix, and data relating to the length of each segment being stored with its corresponding acoustic event.
Claim 12
Dependent Claim
29 words
A method according to claim 10 wherein the step of comparing the time encoded speech symbol stream with the archetypes stored as time encoded speech comprises dynamic time warping.
Claim 13
Dependent Claim
21 words
A method according to claim 11 wherein in the step of comparing, each segment is weighted in dependence upon its duration.
Claim 14
Dependent Claim
23 words
A method according to claim 12 wherein the step of comparing comprises comparing symbol event descriptors with the archetypes of words or utterances.
Claim 15
Dependent Claim
20 words
A method according to claim 13 wherein the symbol event descriptors comprise groups of time encoded symbols, in any combination.
Claim 16
Dependent Claim
18 words
A method according to claim 13 wherein the symbol event descriptors comprise derivatives of the time encoded symbols.
Claim 17
Dependent Claim
29 words
A method according to claim 13 wherein the symbol event descriptors comprise in combination, derivatives of time encoded symbols and time encoded symbols or groups of time encoded symbols.
Claim 18
Dependent Claim
162 words
Apparatus for recognizing voice signals encoded as time encoded speech symbols, the apparatus comprising: receiver means for receiving a time encoded speech symbol stream and segmenting the stream into a number of time frames, a plurality of seeker circuits, arranged to receive the time frames and being optimized for detection of an acoustic event of the voice signals to be recognized, classifying means for examining the time encoded speech symbols in the time frames and for determining the presence or absence of any acoustic event to which any seeker circuit is optimized, segmentation means for determining segmentation boundaries in the time encoded speech symbol stream, means for storing archetypes of words or utterances stored as time encoded speech, comparator means for comparing the time encoded speech symbol stream within the time frames with archetypes of words or utterances stored as time encoded speech, and output means for providing an output signal indicative of the voice signal as a result of the comparison.
Claim 19
Dependent Claim
162 words
Apparatus for recognizing voice signals encoded as time encoded speech symbols, the apparatus comprising: receiver means for receiving a time encoded speech symbol stream and segmenting the stream into a number of time frames, a plurality of seeker circuits, arranged to receive the time frames and being optimized for detection of an acoustic event of the voice signals to be recognized, classifying means for examining the time encoded speech symbols in the time frames and for determining the presence or absence of any acoustic event to which any seeker circuit is optimized, segmentation means for determining segmentation boundaries in the time encoded speech symbol stream, means for storing archetypes of words or utterances stored as time encoded speech, comparator means for comparing the time encoded speech symbol stream within the time frames with archetypes of words or utterances stored as time encoded speech, and output means for providing an output signal indicative of the voice signal as a result of the comparison.
Claim 20
Dependent Claim
33 words
Apparatus according to claim 17, and further comprising processing means operable on the time frames received by the seeker circuits for emphasizing any low frequency contents of the voice signals to be recognized.
Claim 21
Dependent Claim
33 words
Apparatus according to claim 17, and further comprising processing means operable on the time frames received by the seeker circuits for emphasizing any high frequency contents of the voice signals to be recognized.
Claim 22
Dependent Claim
16 words
Apparatus according to claim 17 18 or 19 wherein the seeker circuits comprise integrated circuit devices.
References Cited (Backward Citations)
US3679830A
Cited
EP0054365A1
Cited
EP0119835A1
Cited
US4763278A
Cited
EP0141497A1
Cited
US4783807A
Cited
EP0173986A2
Cited
US4852170A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1995-03-31 | 1995-09-30 | 1996-03-31 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1999-03-31 | 1999-09-30 | 2000-03-31 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2003-03-31 | 2003-09-30 | 2004-03-31 | $7700.00 | Window Closed / Lapsed |