Patent US5890104A successfully refreshed from Google Patents.
US5890104A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method and apparatus for testing telecommunications equipment using a reduced redundancy test signal
Assignee: Psytechnics Ltd
•
Filing: 1994-06-17
•
Granted: 1999-03-30
•
Expires: 2014-06-17
Abstract
Abstract A test signal for a telecommunications apparatus includes a succession of segments of real or synthetic speech which includes different successions of the same sounds; rarely occurring sounds; and sounds at extremes of level, pitch and duration, so as to be similar to natural speech with redundancy removed. In analyzing the performance of the telecommunications apparatus the analyzer takes into account the frequency of occurrence of each sound in natural speech.
Inventors
- 👤 Michael Peter Hollier
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 14:34
CPC Classifications
H; H04; H04M; H04M3/00; H04M3/22; H04M3/2236; G; G10; G10L; G10L25/00; G10L25/48; G10L25/69; H04M1/00; H04M1/24; H04M2201/00; H04M2201/40
Claims Analysis
Total Claims: 27
Claim 1
Independent Claim
131 words
A method of generating a test signal for testing telecommunications apparatus, the test signal comprising: a sequence of segments, each segment reproducing a basic phoneme found in natural speech; wherein the sequence provides representative examples of a variety of basic phonemes and phoneme attributes, including formant structures and temporal structures corresponding to natural speech wherein no phoneme appears with the same attributes more than once and the sequence is generated by selecting speech segments from a store in which each basic phoneme occurring in natural speech is represented; and wherein, if within the sequence there is generated a subsequence of two or more basic phonemes which cannot occur in natural speech for linguistic or physiological reasons, such a subsequence is identified and excluded from forming part of the test signal transmitted.
Claim 2
Independent Claim
131 words
A method of generating a test signal for testing telecommunications apparatus, the test signal comprising: a sequence of segments, each segment reproducing a basic phoneme found in natural speech; wherein the sequence provides representative examples of a variety of basic phonemes and phoneme attributes, including formant structures and temporal structures corresponding to natural speech wherein no phoneme appears with the same attributes more than once and the sequence is generated by selecting speech segments from a store in which each basic phoneme occurring in natural speech is represented; and wherein, if within the sequence there is generated a subsequence of two or more basic phonemes which cannot occur in natural speech for linguistic or physiological reasons, such a subsequence is identified and excluded from forming part of the test signal transmitted.
Claim 3
Dependent Claim
19 words
A test signal as in claim 1, wherein some of the basic phonemes in the sequence comprise voiced fricatives.
Claim 4
Dependent Claim
31 words
A test signal as in claim wherein some of said basis phonemes are present in a plurality of instances, each with different values of an attribute of speech within the sequence.
Claim 5
Dependent Claim
28 words
A test signal as in claim 3 in which one attribute value is an average value, and at least one other is a more rarely occurring attribute value.
Claim 6
Dependent Claim
20 words
A test signal as in claim 3 in which the attribute includes at least one of volume, pitch, and duration.
Claim 7
Dependent Claim
15 words
A test signal as in claim 3 in which the attribute includes a co-articulation effect.
Claim 8
Dependent Claim
15 words
A test signal as in claim 6 in which the attribute includes formant transition rate.
Claim 9
Dependent Claim
29 words
A test signal as in claim 6 in which some of said basic phonemes are present in a plurality of instances, each instance immediately following a different predecessor phoneme.
Claim 10
Dependent Claim
23 words
A test signal as in claim 6 in which smoothed transitions between the characteristics of each segment and its following segment are represented.
Claim 11
Dependent Claim
14 words
A test signal as in claim 1 comprising concatenated segments of prerecorded human speech.
Claim 12
Dependent Claim
31 words
A method as in claim 1 comprising the step of generating, for each segment selected, smoothed transitions between the characteristics of that segment and the immediately preceding segment in the sequence.
Claim 13
Independent Claim
133 words
A method of testing telecommunications apparatus, the method comprising the steps of: supplying thereto a test signal comprising a sequence of basic phonemes corresponding to parts of words found in natural speech, analyzing a signal output from the telecommunications apparatus on each of said segments; and forming from said signal output an overall measure of the distortion caused by the telecommunications apparatus, the test signal providing representative examples of a variety of basic phonemes and phoneme attributes, including formant structures and temporal structures corresponding to natural speech wherein no phoneme appears with the same attributes more than once, and a signal representing the frequency of occurrence in natural speech of each segment; the overall measure of distortion being based on the distortion of each segment and its frequency of occurrence in natural speech.
Claim 14
Independent Claim
133 words
A method of testing telecommunications apparatus, the method comprising the steps of: supplying thereto a test signal comprising a sequence of basic phonemes corresponding to parts of words found in natural speech, analyzing a signal output from the telecommunications apparatus on each of said segments; and forming from said signal output an overall measure of the distortion caused by the telecommunications apparatus, the test signal providing representative examples of a variety of basic phonemes and phoneme attributes, including formant structures and temporal structures corresponding to natural speech wherein no phoneme appears with the same attributes more than once, and a signal representing the frequency of occurrence in natural speech of each segment; the overall measure of distortion being based on the distortion of each segment and its frequency of occurrence in natural speech.
Claim 15
Dependent Claim
31 words
A method as in claim 12 wherein some of said basic phonemes are present in a plurality of instances, each with different values of an attribute of speech within the sequence.
Claim 16
Dependent Claim
27 words
A method as in claim 13 in which one attribute value is an average value, and at least one other is a more rarely occurring attribute value.
Claim 17
Dependent Claim
19 words
A method as in claim 13 in which the attribute includes at least one of volume, pitch and duration.
Claim 18
Dependent Claim
14 words
A method as in claim 13 in which the attribute includes a co-articulation effect.
Claim 19
Dependent Claim
14 words
A method as in claim 16 in which the attribute is formant transition rate.
Claim 20
Dependent Claim
28 words
A method as in claim 16 in which some of said basic segments are present in a plurality of instances, each instance immediately following a different predecessor segment.
Claim 21
Independent Claim
253 words
A method for producing a test signal and testing telecommunications apparatus, the method comprising the steps of: supplying thereto a test signal comprising a sequence of basic phonemes corresponding to parts of words found in natural speech, analyzing a signal output from the telecommunications apparatus on each of said segments; and forming from said signal output an overall measure of the distortion caused by the telecommunications apparatus, the test signal providing representative examples of a variety of basic phonemes and phoneme attributes, including formant structures and temporal structures corresponding to natural speech wherein no phoneme appears with the same attributes more than once and a signal representing the frequency of occurrence in natural speech of each segment; the overall measure of distortion being based on the distortion of each segment and its frequency of occurrence in natural speech; said method including producing a test signal by repeatedly selecting individual speech segments from a superset of such segments to form a subset of predetermined length, wherein the following steps are performed for each speech segment so selected: testing for the presence of the individual segment in the subset more than a predetermined number of times, testing the immediately preceding member of the subset, to determine if the sequence of the selected segment and the immediately preceding segment is excluded by physiological and/or linguistical constraints, if either condition occurs, rejecting the segment and selecting a further segment from the superset; and wherein the resulting sequence is digitally edited so as to remove discontinuities between segments.
Claim 22
Independent Claim
253 words
A method for producing a test signal and testing telecommunications apparatus, the method comprising the steps of: supplying thereto a test signal comprising a sequence of basic phonemes corresponding to parts of words found in natural speech, analyzing a signal output from the telecommunications apparatus on each of said segments; and forming from said signal output an overall measure of the distortion caused by the telecommunications apparatus, the test signal providing representative examples of a variety of basic phonemes and phoneme attributes, including formant structures and temporal structures corresponding to natural speech wherein no phoneme appears with the same attributes more than once and a signal representing the frequency of occurrence in natural speech of each segment; the overall measure of distortion being based on the distortion of each segment and its frequency of occurrence in natural speech; said method including producing a test signal by repeatedly selecting individual speech segments from a superset of such segments to form a subset of predetermined length, wherein the following steps are performed for each speech segment so selected: testing for the presence of the individual segment in the subset more than a predetermined number of times, testing the immediately preceding member of the subset, to determine if the sequence of the selected segment and the immediately preceding segment is excluded by physiological and/or linguistical constraints, if either condition occurs, rejecting the segment and selecting a further segment from the superset; and wherein the resulting sequence is digitally edited so as to remove discontinuities between segments.
Claim 23
Dependent Claim
165 words
Test apparatus for testing telecommunications apparatus, the test apparatus comprising: means for supplying a test signal to the apparatus under test, the test signal comprising a sequence of basic phonemes corresponding to parts of words found in natural speech, means for returning the signal, as modified by the apparatus under test, to the test apparatus, and means for analyzing the signal output from the telecommunications apparatus on each of said segments; the means for supplying the test signal being arranged to supply a test signal in which the segments are assembled into a sequence which provides representative examples of a variety of basic phonemes and phoneme attributes, and which includes a formant structure and temporal structure corresponding to natural speech wherein no phoneme appears with the same attributes more than once, and the analysis means including means for forming an overall measure of the distortion caused by the telecommunications apparatus based on the distortion of each segment and its frequency of occurrence in natural speech.
Claim 24
Dependent Claim
165 words
Test apparatus for testing telecommunications apparatus, the test apparatus comprising: means for supplying a test signal to the apparatus under test, the test signal comprising a sequence of basic phonemes corresponding to parts of words found in natural speech, means for returning the signal, as modified by the apparatus under test, to the test apparatus, and means for analyzing the signal output from the telecommunications apparatus on each of said segments; the means for supplying the test signal being arranged to supply a test signal in which the segments are assembled into a sequence which provides representative examples of a variety of basic phonemes and phoneme attributes, and which includes a formant structure and temporal structure corresponding to natural speech wherein no phoneme appears with the same attributes more than once, and the analysis means including means for forming an overall measure of the distortion caused by the telecommunications apparatus based on the distortion of each segment and its frequency of occurrence in natural speech.
Claim 25
Dependent Claim
54 words
Apparatus as in claim 20, wherein the test signal supply means is arranged to supply a test signal comprising a sequence of basic phonemes corresponding to parts of words found in natural speech, some of said basic phonemes being present in a plurality of instances, but with different values of an attribute of speech.
Claim 26
Independent Claim
89 words
A method of testing telecommunications apparatus, said method comprising: supplying to said apparatus a test signal comprising a plurality of signal segments corresponding to speech like sounds, in which the represented sounds which occur relatively rarely in natural speech have a higher relative frequency of occurrence; analysing the signal output from the telecommunications apparatus on each of said sounds; and forming an overall measure of the distortion caused by the telecommunications apparatus based on the distortion of each signal segment sound and its frequency of occurrence in natural speech.
Claim 27
Independent Claim
89 words
A method of testing telecommunications apparatus, said method comprising: supplying to said apparatus a test signal comprising a plurality of signal segments corresponding to speech like sounds, in which the represented sounds which occur relatively rarely in natural speech have a higher relative frequency of occurrence; analysing the signal output from the telecommunications apparatus on each of said sounds; and forming an overall measure of the distortion caused by the telecommunications apparatus based on the distortion of each signal segment sound and its frequency of occurrence in natural speech.
References Cited (Backward Citations)
US4352182A
Cited
DE3032699A1
Cited
US4972484A
Cited
US4860360A
Cited
GB2218300A
Cited
GB2218299A
Cited
US5195124A
Cited
US5369644A
Cited
WO1994000922A1
Cited
US5490234A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2002-03-29 | 2002-09-28 | 2003-03-30 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2006-03-29 | 2006-09-28 | 2007-03-30 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2010-03-29 | 2010-09-28 | 2011-03-30 | $7700.00 | Window Closed / Lapsed |