EP0240330A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Noise compensation in speech recognition
Assignee: National Research Development Corp UK
•
Filing: 1987-04-01
•
Granted: —
•
Expires: 2007-04-01
Abstract
Abstract In speech recognition it is advantageous to take account of noise levels both in recognition and training. In both processes signals reaching a microphone l0 are digitised and passed through a filter bank to be separated into frequency channels. In training, a noise estimator 20 and a masker l5 are used with a recognizer l8 to prepare and store probability density functions (p.d.f.s) for each channel partially defining Markov models of words to be recognized. The p.d.f.s are derived only from input signals above noise levels but derivation is such that the whole of each p.d.f. is represented. In recognition, "distance" measurements on which recognition is based are derived for each channel. If the signal in a channel is above noise then the distance is determined. by the recognizer, from the negative logarithm of the p.d.f. but if a channel signal is below noise then the distance is determined from the negative logarithm of the cumulative distance of the p.d.f. to the noise level.
Inventors
- 👤 Nigel Charles Sedgwick
- 👤 John Nicholas Holmes
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-10-29 09:35
CPC Classifications
G; G10; G10L; G10L15/00; G10L15/20
Claims Analysis
Total Claims: 19
Claim 1
Independent Claim
180 words
Apparatus for use in sound recognition comprising means for deriving a plurality of input signals during recognition which are each representative of signal levels in respective regions in the frequency spectrum, means for storing a plurality of groups of p.d.f. values representing probability density functions indicating the likelihoods that input signals arise from states in finite state machine models of groups of sounds to be recognized, means for estimating the input noise level, and means for recognizing sounds from the input signals, the stored p.d.f. values and the models, employing respective distance measures, each derived from one input signal and one p.d.f. as represented by one group of the said values, characterised in that each distance measure represents a likelihood of obtaining a region signal level from one p.d.f. when the input signal is above a predetermined level related to the noise level in the corresponding spectrum region, and represents the cumulative likelihood of obtaining from the said p.d.f. a region signal level below the said predetermined level, when the input signal is below, or equal to, the predetermined level.
Claim 2
Dependent Claim
180 words
Apparatus for use in sound recognition comprising means for deriving a plurality of input signals during recognition which are each representative of signal levels in respective regions in the frequency spectrum, means for storing a plurality of groups of p.d.f. values representing probability density functions indicating the likelihoods that input signals arise from states in finite state machine models of groups of sounds to be recognized, means for estimating the input noise level, and means for recognizing sounds from the input signals, the stored p.d.f. values and the models, employing respective distance measures, each derived from one input signal and one p.d.f. as represented by one group of the said values, characterised in that each distance measure represents a likelihood of obtaining a region signal level from one p.d.f. when the input signal is above a predetermined level related to the noise level in the corresponding spectrum region, and represents the cumulative likelihood of obtaining from the said p.d.f. a region signal level below the said predetermined level, when the input signal is below, or equal to, the predetermined level.
Claim 3
Independent Claim
176 words
A method for use in sound recognition comprising the steps of deriving a plurality of input signals during recognition which are each representative of signal levels in respective regions in the frequency spectrum, storing a plurality of groups of p.d.f. values representing probability density functions indicating the likelihoods that input signals arise from states in finite state machine models of groups of sounds to be recognized, estimating the input noise level, and recognizing sounds from the input signals, the stored p.d.f. values and the models, employing respective distance measures, each derived from one input signal and one p.d.f. as represented by one group of the said values, characterised in that each distance measure represents a likelihood of obtaining a region signal level from one p.d.f. when the input signal is above a predetermined level related to the noise level in the corresponding spectrum region, and represents the cumulative likelihood of obtaining from the said p.d.f. a region signal level below the said predetermined level, when the input signal is below, or equal to, the predetermined level.
Claim 4
Independent Claim
176 words
A method for use in sound recognition comprising the steps of deriving a plurality of input signals during recognition which are each representative of signal levels in respective regions in the frequency spectrum, storing a plurality of groups of p.d.f. values representing probability density functions indicating the likelihoods that input signals arise from states in finite state machine models of groups of sounds to be recognized, estimating the input noise level, and recognizing sounds from the input signals, the stored p.d.f. values and the models, employing respective distance measures, each derived from one input signal and one p.d.f. as represented by one group of the said values, characterised in that each distance measure represents a likelihood of obtaining a region signal level from one p.d.f. when the input signal is above a predetermined level related to the noise level in the corresponding spectrum region, and represents the cumulative likelihood of obtaining from the said p.d.f. a region signal level below the said predetermined level, when the input signal is below, or equal to, the predetermined level.
Claim 5
Dependent Claim
29 words
Apparatus or method according to Claim l or 2 characterised in that the groups of sounds are words and the regions are channels defined by a bank of filters.
Claim 6
Dependent Claim
49 words
Apparatus according to Claim l or 3 characterised in that the means for recognizing sounds comprises means for deriving masked input signals during recognition by representing any channel of an input signal below noise level with a masking level which is representative of the noise level in that channel.
Claim 7
Dependent Claim
45 words
Apparatus according to Claim l, 3 or 4 characterised in that the means for estimating the noise level includes a first microphone for sensing noise and the means for deriving input signals includes a second separate microphone for sensing noise plus sounds to be recognized.
Claim 8
Dependent Claim
42 words
Apparatus according to Claim l, 3 or 4 characterised in that the means for deriving input signals and the means for estimating noise level include a common microphone, and means for differentiating between noise only and noise plus sounds to be recognized.
Claim 9
Dependent Claim
88 words
Apparatus or method according to any preceding claim characterised in that each said distance measure is derived from when the noise level is below the input signal and from -ln[erf((A-m)/s)] when the noise level is above the input signal; where A is the noise level in the spectrum region corresponding to the input signal, f is the input level in a region due to sounds to be recognized, and m and s are a mean and a variance representing the p.d.f. from which that distance measure is derived.
Claim 10
Independent Claim
121 words
A method of training a sound recognition system comprising deriving a plurality of groups of input signals from repetitions of nominally the same sound, each group being representative of signal levels in respective regions in the frequency spectrum, and deriving a plurality of groups of p.d.f. values representing probability density functions indicating the likelihoods that input signals arise from states in finite state machine models for a vocabulary of groups of sounds to be recognized, characterised in that the p.d.f. values are derived only from input signals above the noise levels in corresponding spectrum regions, and the derivation is so carried out that the groups of values represent substantially whole probability functions although obtained from input signals above noise levels only.
Claim 11
Independent Claim
121 words
A method of training a sound recognition system comprising deriving a plurality of groups of input signals from repetitions of nominally the same sound, each group being representative of signal levels in respective regions in the frequency spectrum, and deriving a plurality of groups of p.d.f. values representing probability density functions indicating the likelihoods that input signals arise from states in finite state machine models for a vocabulary of groups of sounds to be recognized, characterised in that the p.d.f. values are derived only from input signals above the noise levels in corresponding spectrum regions, and the derivation is so carried out that the groups of values represent substantially whole probability functions although obtained from input signals above noise levels only.
Claim 12
Dependent Claim
42 words
A method according to Claim 8 characterised in that the noise level used in each region of the frequency spectrum is the highest found in deriving the input signals for that region for all training repetitions of all sounds in the vocabulary.
Claim 13
Dependent Claim
40 words
l0. A method according to Claim 8 or 9 characterised in that each p.d.f. is assumed to have a normal distribution and each group of p.d.f. values comprises the true mean m and the true variance s₂ of the distribution.
Claim 14
Independent Claim
50 words
A method according to Claim l0 characterised in that m and s are determined from where B is noise level, M is the mean of samples above the noise level, F is the proportion of input signals below the noise level, erf(F) is as defined above, and Q(F) = N(erf⁻¹(F),0,l).
Claim 15
Independent Claim
51 words
A method according to Claim l0 or ll characterised in that a constant mean and a constant variance are substituted for the said values in any said group representing a p.d.f. derived in training in which the proportion of input signals which are noise exceeds a predetermined value greater than 0.5.
Claim 16
Independent Claim
14 words
A method according to Claim l2 characterised in that the predetermined value is 0.8.
Claim 17
Independent Claim
40 words
A method according to Claim l2 characterised in that if the said proportion is between the predetermined value and a lower further predetermined value then a smooth transition of values is arranged over the range between the two predetermined values.
Claim 18
Independent Claim
39 words
A method according to any of Claims l0 to l4 characterised in that a standard minimum variance is added to all calculated variances, the standard minimum variance having a value such that the variances for different states differ significantly.
Claim 19
Independent Claim
51 words
A method according to any of Claims l0 to l6 characterised in that in deriving the said groups of values. the standard minimum variance for a particular p.d.f. is scaled by a function of the number of input signal samples used to derive the said group of values for that p.d.f.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.