Device for measuring maximal legibility of speech

FIELD: technology for measuring maximally possible legibility of speech in extremely unfavorable conditions, in noise, with low signal/noise ratios and is meant, primarily, for determining protection of objects in case of speech information loss through several leaking channels simultaneously.

SUBSTANCE: device contains generator of acoustic testing signal in form of series of N frequencies with pauses between frequencies, emitter, acoustic signal receiver, N-band device for measuring signal/noise rations and device for computing legibility. In addition, in parallel to receiver of acoustic signal K receivers of other signal types, formed by emission of acoustic testing signal, are mounted, for example, for detecting magnetic, electronic, vibro-acoustic types, while clock generator has additional outputs for organization of K+1 cycles of generator and synchronous control over switch of receivers and selection device.

EFFECT: increased trustworthiness of speech legibility measurements.

2 dwg

 

The technical field.

The invention relates to speech recognition, specifically to the measurement of maximum speech intelligibility under extremely adverse conditions, the noise at low signal to noise, and is intended primarily to determine the degree of protection against leakage of voice data across multiple leak paths simultaneously, i.e. in real terms.

The level of technology.

There are known various devices for measuring the intelligibility of speech. Popular expert listening in place of receiving acoustic signals, composed of specially selected phrases, words or syllables (programs)that are emitted in the place of generation of the speech, with subsequent processing of the results to determine the intelligibility (Pampelmuse, Addchance. Electro-acoustic paths with feedback. - M.: Communication, 1969; GOST 71-53-85 "Measuring the intelligibility of speech or touchtone method"; GOST In 20775-75. "The voice transmission over communication paths, equipped with privacy. Requirements for speech intelligibility. Method of measurement"). These papers described a similar measurement devices used in this equipment and methods of mathematical processing. The device is technically simple, but the measurements are not sufficiently reproducible and accurate in terms of weak acoustic signals, since it depends on the intelligence of the experts, and, most important, time consuming and involve a large number of experts.

A more objective method and devices based on coherent acoustic radiation of separate tonal components of the audible spectrum in the transmission and determining any experts of their threshold of audibility in the receive location, and actually receive location can only be microphones connected via amplifiers with headphones experts (Nebroski. Calculation and measurement of speech intelligibility. - M.: Communication, 1962). The acoustic radiation level of each component is increased to the threshold of hearing her an expert. Recorded radiation levels mathematically processed taking into account the curve of human audibility and thus determine the minimum level of the speech signal to obtain a given speech intelligibility at the receiving location. The information, of course, can also be used to determine the maximum level of the acoustic signal, wherein in the predetermined acceptance speech is guaranteed to be legible, i.e. provided with a certain level of protection against leakage of acoustic information. However, this method is time-consuming, because it requires the participation of many experts and time. In addition, the estimate of intelligibility is not on the real level of acoustic signals, and those that are appropriate to lstout the threshold of hearing at the specified location. This is often unacceptable, because, as you know, the parasitic channels of transmission of acoustic information (walls, floors, Windows, furniture, doors etc) usually unevenly attenuate frequency components of speech and the protection of information from leakage, calculated according to the results of the above measurements may not correspond to actual.

Currently, the most common instrumental methods based on the tonal method of speech intelligibility (e.g., Pampelmousse, Addchance. Electro-acoustic paths with feedback. - M.: Communication, 1969, GOST In 20775-75, Useful model # 27259). They provide a measure of intelligibility in any ratio signal/noise purely technical means, without the use of expert listening. Most modern device patent for useful model №27259, publ. 10.01.2003. It is the closest to the claimed device on the most essential features and, therefore, selected as a prototype.

Below is a detailed description of the device the prototype.

It is illustrated by figure 1 - example of a simplified structural diagram, where: 1 - generator N frequency test signal, 2 - acoustic emitter (speaker), 3 - acoustic receiver (microphone), 4 - N-band meter signal to noise ratio, 5 - evaluator level of speech intelligibility.

Give the TWT consists of series-connected generator 1 of the test signal, the loudspeaker 2, placed in the place of the actual location of the sound source signals, a microphone 3, placed at the listening position, N-band meter 4 signal to noise ratio on each of the N frequencies of the test signals and the transmitter are 5 levels of speech understanding.

The device operates as follows. After placement of the speaker 2 and the microphone 3 in these locations include the device. Generator 1 generates an acoustic test signal to the specified level, consisting of a sequence of N frequencies, separated by medium frequency N bands, which are divided audible range of speech. Speaker 2 emits the specified signals. Microphone 3 converts the signals received at the specified radiation and in the pauses between them) into electricity. While paused, the same receiver receive and measure the signal noise in this band. According to the results of measurements determine the power value of a received signal of this tonal frequency, on the basis of independent linear summation of the capacities of the signal and noise determine signal/noise ratio on this tone frequency. Then the procedure is repeated on the next tone frequency, then the next, until you have defined the relation signal/noise at all tone frequencies of the test signal. This occurs in the N-band meter 4 relations C is the channel noise. Thus, at the input of the transmitter 5 level of intelligibility in turn receives the measured ratio of signal to noise for all N emitted frequencies. The transmitter 5 level of intelligibility on the measured signal to noise makes calculation of speech intelligibility. The calculation is made on the basis of known relationships and theories, the essence of which is that initially above the measured signal to determine the noise intelligibility coefficients for each frequency of the test signal. In certain respects, the signal/noise at all tone frequencies of the test signal to calculate the coefficient of intelligibility on the real background noise on the basis of known techniques (see, for example, article Ana, Wranovics, Vchgetdate. Correlation theory of speech intelligibility in the journal "electronics", series "General electronics", 1995, issue 1, p.3-9; Nebroski. Calculation and measurement of speech intelligibility. - M.: Communication, 1962).

Them determine the average (arithmetic mean) value of the coefficient of intelligibility. Then at a certain average value of the coefficient of intelligibility and known curve verbal intelligibility from the value of the coefficient of intelligibility define verbal intelligibility of speech (see, for example, in stat is e Vchgetdate and other "Some methodological approaches to the evaluation of protective efficacy of verbal information" in the magazine "Special technique", 2000, No. 4, p.39-45).

In other words, the essence of the device, and other devices for measuring the intelligibility is coherent radiation at the point of transfer test N tonal acoustic oscillations in the speech frequency range, the acoustic received signals at a given point of reception, the N-band measure of signal to noise for all N frequencies of the test signal and the subsequent calculation of the coefficient of intelligibility - known methods.

The described devices are widely used for measuring the intelligibility of speech, i.e. to determine the quality of listening. Currently, however, no less significant problem was the fight against eavesdropping, i.e. reducing the legibility in places of possible information leakage, prevent unauthorized listening. In this case (unlike the problem of determining the quality of listening) the ratio signal/noise is small and the attacker is difficult to parse the speech signal. As is known, in such cases, eavesdropping use not only the acoustic, but also all other possible channels of information leakage, catching and processing, in particular, magnetic fields, to make the sound equipment, and vibro-acoustic fluctuations associated with its work. The signals of these channels may in some cases be more legible than acoustic. In such cases, the application of the described device for controlling the intelligibility of speech is not acceptable, it does not reflect possible level of speech intelligibility.

Naturally, the above-described device for measuring acoustic intelligibility can be modified and adapted for measuring the intelligibility of speech over other possible channels of information leakage, the most significant of which are magnetic, electric and acoustic vibration channels. Then we can estimate the maximum level of intelligibility, as the greatest of all these channels. However, these devices cannot measure the real level of protection of verbal information from unauthorized eavesdropping, because when you use to listen to recordings of signals of all possible leak paths possible further joint processing of these records to extract speech information (see, for example, Aghani, Ali and other Robustness of information systems management: a Training manual. (In MAP "World book", 1993, which describes the optimal discriminator two deterministic signals, the maximum likelihood estimation of the parameter signal is with the initial phase, optimal signal processing for models of the useful signal with a random initial phase in the form of a quadrature signal processing. This may allow you to get the level of intelligibility, greater than the maximum in any of the individual channels. This possibility stems from the fact that, as you know, each channel of information leakage castotnozawisimaya. Accordingly, the frequency dependence of the signal to noise of different channels. The ratio signal/noise will be different in the same frequency band by different channels of leakage. Based on this hypothetical is probably the most sensitive, the most rational improve the intelligibility of speech, providing separation was obtained for analysis of the signals from each channel into several frequency bands and the enumeration of all possible combinations of signals of the frequency bands of the channels between them for superior intelligibility. The implementation of such eavesdropping devices, difficult road, but nevertheless it is possible.

For the evaluation of protection for sound amplification equipment in such a multi-method eavesdropping (which, to our knowledge, yet nowhere hardware is not implemented, but is quite feasible) required methods of measurement for maximum intelligibility, taking into consideration its limits. We called it the dimension maximum intelligibility".

Proceed. of the local device the prototype of such measurements can not provide.

The objective of the invention.

The present invention is to increase the reliability of the measurement of speech intelligibility in multiple channels of leakage of verbal information.

The solution is the essence of the invention.

The problem is solved by the fact that in the known device for the measurement of speech intelligibility, containing connected in series generator acoustic test signal as a sequence of N frequencies distributed according to the average frequency bands, which are divided audible range, with pauses between frequencies, the emitter, the receiver of an acoustic signal, the N-band meter signal to noise and the transmitter intelligibility, and the generator and N-band meter are synchronously controlled clock synchronizer switches bandwidths, made significant changes and additions, namely:

parallel to the acoustic signal receiver mounted To the receivers of other types of signals formed by the emission of an acoustic test signal, for example, magnetic, electric, vibroacoustic,

- at the point of reception is additionally place To receivers of other types of signals generated acoustic test signal, such as electrical, magnetic, vibroacoustic signal is Alov,

each frequency N frequency test signal and pause repeated cyclically To+1 times.

the output of the acoustic receiver and other receivers of the signals connected to the input of the N-band meter signal/noise through the switch receivers are synchronized with the beginning of each of the K+1 loop repeat,

output N-frequency measuring signal to noise connected to the input of the transmitter intelligibility through the device selection in each cycle the highest values of signal to noise for each frequency test signal,

when this clock generator has additional outputs for looping generator control switch receiver and device selection.

Disclosure of the invention.

The essence of the invention is to provide measurements of intelligibility when the largest of all types of received signals ratios of signal to noise in each frequency band. When unauthorized eavesdropping conditions for obtaining maximum intelligibility, even in theory, may not be the best. Thus, we define the maximum speech intelligibility.

An example of a simplified structural diagram of the device represented in figure 1, where the following notation: 1 - generator N-frequency test signal, 2 - acoustic emitter (speaker), 3 - pickup the nick of acoustic signals (microphone), 4 - N-band meter signal to noise ratio, 5 - evaluator level of intelligibility, 6 - magnetic signals, 7 - receiver of electrical signals, 8 - receiver vibroacoustic signals, 9 - switch receivers, 10 - device selecting the highest signal-to-noise ratio, 11 - clock synchronizer. In the example, the number of other types of receivers, K=3 (magnetic, electric and acoustic vibration).

The device consists of series-connected generator 1 N-frequency test signal, the loudspeaker 2, placed in the place of the actual location of the source of speech signals, a microphone 3, the receivers 6, 7, 8 magnetic, electric and vibroacoustic signals, placed at the listening position, the switch 9 of receivers, N-band meter 4 signal to noise, the device 10 selecting the highest signal-to-noise and transmitter 5 levels of speech understanding.

The device is synchronized such synchronizer 11.

First place the source of the acoustic test signal (speaker 2) at the point of placement of the source speech signal, and receiver 3 acoustic, magnetic, 6, 7 electric and vibroacoustic 8 signals at the receiving end of verbal information.

Generator 1 generates, emitter 2 emits the test tone back the level of, which are, most often, the sequence of pitch frequencies distributed according to the average frequency N bands, which are divided audible range, with pauses between frequencies. This radiation and pause at each frequency cycle is repeated K+1=4 times.

Switching in each cycle the switch 9 receivers 2, 6, 7, 8 alternately, accept and N-band meter 4, measured at the point of reception of the received acoustic, magnetic, electric and acoustic vibration signals in the radiation and in the pauses between them. The received signals are separated by N narrow-band filters at frequencies corresponding to the frequencies of the test signal. Narrowband filters are switched by synchronizer 11 simultaneously with the change of frequency of generation of the test signal. Because measured levels of the received signals at the output of each receiver (in our example, acoustic, magnetic, electric and acoustic vibration)define four signal-to-noise ratio for each frequency of the test signal. Then the device 9 selects at each frequency of the test signal (i.e. each band of the audible spectrum with the highest signal-to-noise ratio. The operation of the switch 9 receivers and the device 10 selecting the highest signal-to-noise synchronized clock synchronizer 11. Thus, the input of the calc is slitely 5 level of intelligibility in turn receives the highest signal/noise for all N emitted frequencies. The transmitter 5 level of intelligibility on the measured signal to noise makes calculation of speech intelligibility. The calculation is made on the basis of known relationships and theories, the essence of which is that initially above the measured signal to determine the noise intelligibility coefficients for each frequency of the test signal. Them determine the average (arithmetic mean) value of the coefficient of intelligibility. Then at a certain average value of the coefficient of intelligibility and known curve verbal intelligibility from the value of the coefficient of intelligibility determine the maximum verbal intelligibility of speech (see, for example, article V. ironstone and other "Some methodological approaches to the evaluation of protective efficacy of verbal information" in the magazine "Special technique", 2000, No. 4, p.39-45).

The main differences of the claimed device from the prototype:

parallel to the acoustic signal receiver mounted To the receivers of other types of signals generated by the acoustic radiation, for example, magnetic, electric, vibroacoustics,

each frequency sequence N-frequency test signal and pause repeated cyclically K+1 times,

the output of the acoustic receiver and the receivers of other signals is otklucheny to the input of the N-band meter signal/noise through the switch receivers, toggle synchronizer at the beginning of each of the K+1 cycle

output N-frequency measuring signal to noise connected to the input of the transmitter intelligibility through the device selection in each cycle (i.e. for each frequency test signal) of the largest value of the ratio signal/noise

when this clock generator has additional outputs for looping generator control switch receiver and device selection.

Such a device is unknown from available sources of information the Combination of the above differences is creative achievement by the inventor, it is not obvious for the expert.

Industrial applicability.

The claimed device can be easily done as a circuit, and software. Circuit - means, well-developed industry, which follows from the above description of the example execution.

In particular, the receiver of acoustic signals may be a microphone, magnetic signals from a magnetic antenna (see, for example, the patent of the Russian Federation 2152624 "measuring the tension of a magnetic component of an alternating electromagnetic field, an electric signal is an electric antenna (see, for example, the patent of the Russian Federation 2152623 "measuring the strength of the electric component of the alternating electromagnetic field", vibroacoustic signals - piezoelectronicpiezomotors (see, for example, Ikkalasida and other fundamentals of hydro-acoustics and sonar station.- Leningrad:Sudostroenie,1970) Selecting the highest value of signal to noise can be performed, for example, on the basis of the scheme or schemes of remembering.

Other components of the device - generate sequence of N test signals, the emitter, the N-band meter signal to noise, the evaluator level of intelligibility is repeatedly implemented in known devices for measuring the intelligibility (see, for example, testimony 27259 for a utility model publ. 10.01.2003 in bull. No. 1 for 2003)

The claimed device is implemented on the FSUE "Informacoustica successfully passed Metrology and certification tests and is already used in the measurements, the maximum speech intelligibility. Any practical difficulties associated with the use of the invention, is not met.

Actually received benefits before the prototype consist in increasing the accuracy of the level measurement of speech intelligibility. Measured guarantee really potentially maximum intelligibility, taking into account the use of a listening face all the possible leakage of verbal information. In particular, maximum intelligibility measured in the testing process when used in addition to the acoustic channel only is about one magnetic channel, was higher than 5 dB than measured by the method of the prototype. This error is known measuring devices is invalid.

The achieved accuracy of the determination of speech intelligibility enables you to securely install the absence of channels of information leakage, because intelligibility is determined based on the use of all possible leakage paths and the most efficient processing of their signals.

Thus, in our opinion, the claimed technical solution meets all the criteria applicable to the invention, - it is new, not obvious and industrially applicable.

The measurement device maximum intelligibility, containing connected in series generator acoustic test signal as a sequence of N frequencies distributed according to the average frequency bands, which are divided audible range, with pauses between frequencies, the emitter, the receiver of an acoustic signal, the N-band meter signal to noise and the transmitter intelligibility, and the generator and N-band meter are synchronously controlled clock synchronizer switches bandwidths, characterized in that parallel to the acoustic signal receiver mounted To the receivers of other types of signals formed by the emission of an acoustic test signal, for example, m is gnity, electric, acoustic vibration, each frequency sequence N-frequency test signal and pause repeated cyclically K+1 times, the outputs of the acoustic receiver and other receivers of the signals connected to the input of the N-band meter signal/noise through the switch receivers are synchronized with the beginning of each of the K+1 loop repeat test frequency, the output of the N-frequency measuring signal to noise connected to the input of the transmitter intelligibility through the device selection in each cycle the highest values of signal to noise, while the oscillator has additional outputs for the organization To+1 cycles of the generator and synchronous control switch receiver and device selection.



 

Same patents:

FIELD: technology for measuring maximally possible legibility of speech under extremely unfavorable conditions, in the noise, for low signal/noise ratios, and is meant mainly for determining protection of objects during loss of speech information through several leaking channels simultaneously.

SUBSTANCE: in accordance to method, in receipt point K receivers of other signal types, formed by acoustic testing signals, for example, electric, magnetic, vibro-acoustic signals, are positioned, acoustic and aforementioned K signal types formed in receipt position at all N frequencies of testing signal and in pauses between same, are received and measured separately, on basis of results of measurements signal/noise ratios are determined at each frequency of testing signal by each one of acoustic and K other types of received signals, at each frequency of testing signal greatest one of signal/noise relations are selected, measured by acoustic and K other types of received signals, then - speech legibility grade is calculated on basis of greatest signal/noise ratios.

EFFECT: increased trustworthiness of speech legibility measurements.

2 dwg

FIELD: hydro-acoustics, possible use for express estimation of characteristic of direction of hydro-acoustic emitter.

SUBSTANCE: in accordance to first variant of manufacture, in front of hydro-acoustic emitter, hydro-acoustic receiver is positioned, made in form of grid of sound-sensitive elements. The latter is in its turn made in form of object filament coil of wave-optical interference meter, supporting filament coil of which is mounted also in liquid, but outside the emission field of hydro-acoustic emitter. Also, device additionally has position indicator of hydro-acoustic emitter and electronic circuit for separating maximal input signal with increase in distance between hydro-acoustic receiver and hydro-acoustic emitter. It occurs during filling of all cells of hydro-acoustic receiver with radiation field of hydro-acoustic emitter. Knowing the theory of experiment, estimation of opening angle of main petal of characteristic of direction of hydro-acoustic emitter is performed. In accordance to second variant of device manufacture, in additionally includes hydro-acoustic receiver scanner in grid plane. When at output of interference meter signal appears with scanning frequency of hydro-acoustic receiver in process of approach of hydro-acoustic emitter and hydro-acoustic receiver, estimation of thin structure of characteristic of direction of hydro-acoustic emitter is performed, i.e. presence of additional petals within complete characteristic of direction of hydro-acoustic emitter.

EFFECT: removed flaws of prototype, simplified construction.

2 cl, 4 dwg

FIELD: hydroacoustics.

SUBSTANCE: device has hydro-chamber, emitters, hydro-phones and device for tugging and holding calibrated antenna inside hydro-chamber. Emitters and hydro-phones are mounted in hydro-chamber walls oppositely to one another. Hydro-chamber and device for tugging and holding antenna are made in form of separate sections, engaged with one another as a whole unit, and calibrated antenna is electrically and mechanically disconnected from system for reproducing and controlling parameters of hydro-acoustic field (emitters and hydro-phones). Additionally plant has set-point generator, used for generation of signals for emitters.

EFFECT: higher precision, higher efficiency.

8 cl, 6 dwg

FIELD: instrumentation engineering.

SUBSTANCE: proposed method includes reception of source noise signal and generation of control signal as function of relative motion parameters of source and receiver prior to measuring waveform and energy width of spectral digital noise component of source noise signal, this control signal being used to convert time scale of noise signal received. In this way generation of control signal and its purposeful change effected by means of time scale of noise signal received enable reduction of distortions in this signal caused by relative motion of noise source and receiver.

EFFECT: enhanced measurement accuracy.

1 cl, 6 dwg

The invention relates to a measurement technique, namely the measurement of parameters of piezoelectric transducers

The invention relates to the field of hydro location.

The invention relates to measurement techniques, in particular to a device for the calibration of hydrophones

The invention relates to a power measurement of the ultrasonic radiation in the water, and biological environments and can be used in industry and medicine

The invention relates to measuring equipment and Metrology and can be used for calibration and calibration of measurement systems, in particular sonar and hydrophysical converters

FIELD: technology for measuring maximally possible legibility of speech under extremely unfavorable conditions, in the noise, for low signal/noise ratios, and is meant mainly for determining protection of objects during loss of speech information through several leaking channels simultaneously.

SUBSTANCE: in accordance to method, in receipt point K receivers of other signal types, formed by acoustic testing signals, for example, electric, magnetic, vibro-acoustic signals, are positioned, acoustic and aforementioned K signal types formed in receipt position at all N frequencies of testing signal and in pauses between same, are received and measured separately, on basis of results of measurements signal/noise ratios are determined at each frequency of testing signal by each one of acoustic and K other types of received signals, at each frequency of testing signal greatest one of signal/noise relations are selected, measured by acoustic and K other types of received signals, then - speech legibility grade is calculated on basis of greatest signal/noise ratios.

EFFECT: increased trustworthiness of speech legibility measurements.

2 dwg

FIELD: technology for recognition of speech, in particular, method and device for dynamic adjustment of beam in Viterbi search.

SUBSTANCE: method includes selecting beginning width of beam, finding out whether value of probability for frame changes, dynamic adjustment of width of beam, decoding of input speech signal, while width of beam is dynamically adjusted. Also proposed is device including processor, component for speech recognition, connected to processor, memory, connected to processor. Component for speech recognition automatically adjusts width of beam to decode input speech signal.

EFFECT: increased speech recognition reliability.

6 cl, 6 dwg, 4 tbl

FIELD: technology for analyzing speech under unfavorable environmental conditions.

SUBSTANCE: during transformation of spoken command first circular buffer is continuously filled with digitized signal, comb of recursive filters is applied to multiply loosened signal and spectral components are utilized to fill second circular buffer, limits of speech fragment are determined within it on basis of adaptive estimate of noise environment, spectral components of speech fragment are transferred to linear analysis buffer, shortened sign space is received from aforementioned buffer and produced spectral components are compared to standard vectors of database commands.

EFFECT: utilization of device under conditions of, for example, moving vehicle or mechanical industry with high noise pollution level provides for stable recognition of commands independently on particularities of narrators pronunciation, decreased memory volume.

7 cl, 2 dwg

FIELD: engineering of informative system with speech interaction system.

SUBSTANCE: interaction system is connected to informative system and to recognition system. Interaction of user with informative system is performed by using user phrases, processed by recognition system. To achieve the result, system contains application sphere module, meant for setting phrase settings in it, which are constructed on basis of classes of objects, classes of attributes and classes of actions, which are common for aforementioned systems of subject application area and which should be matched by certain types of objects, types of actions, types of attributes and their instances, generated by informative system, for construction of grammar of phrases inputted by user.

EFFECT: possible interaction of user with informative system.

8 cl, 1 dwg

FIELD: analysis and recognition of speech signals, can be used for recognition of speech patterns.

SUBSTANCE: device for realization of aforementioned speech phoneme recognition method has: computing system, including clock generator, controller, random-access memory device, central microprocessor unit, meant for forming bispectral signs and recognizing them on basis of speech phonemes, digital-analog converter, long-term memorizing device, video-controller and analog-digital converter, and also keyboard, display, headphones and a microphone.

EFFECT: increased precision of speech patterns recognition due to forming of phoneme signs for speech phonemes recognition based on application of bispectral analysis, based on transformation of digital code series, appropriate for speech signals, to bispectral zone, characterizing interaction between values of Fourier components at different frequencies within speech spectrum, and thus, to provide selection of an additional, significantly new information from speech signals, to increase precision of phoneme recognition.

2 cl, 5 dwg

FIELD: radio engineering.

SUBSTANCE: device has block for determining beginning and end of command, first memory block, block for syllable segmentation, block for time normalization of command, standard commands block, commands likeness calculator, while output of block for determining beginning and end of command is connected to first inputs of first memory block and syllable segmentation block, output of first memory block is connected to first output of command time normalization block, second output of which is connected to output of syllable segmentation block. Device additionally has supporting noise input, second memory block, block for time normalization of noise, first and second blocks for level normalization, signals mixer, while input of speech command is connected to output of block for determining beginning and end of command and to second inputs of syllable segmentation block and first memory block, bearing noise input is connected to first input of second memory block, to second input of which output of block for determining beginning and end of command is connected, output of second memory block is connected to first input of block for time normalization of noise, output of syllable segmentation block is connected to second inputs of block for time normalization of noise, of first and second level normalization block, standard commands block, and to third inputs of first and second memory block, output of block for time normalization of noise is connected to first input of signals mixer, to second input of which output of standard commands block is connected, first input of which is connected to first output of commands likeness calculator, output of signals mixer is connected to first input of second level normalization block, output of which is connected to second input of commands likeness calculator, to first input of which output of first level normalization block is connected, first input of which is connected to output of block for time normalization of command.

EFFECT: higher probability of correct command recognition during effect from noises.

6 dwg

FIELD: automatic voice recognition technologies.

SUBSTANCE: acoustic signal, observed at a point of person body, unknown to unauthorized personnel, is inputted to computing device, values of parameters of acoustic signal are determined, values of estimates of statistical characteristics of parameters of acoustic signal are determined and standards are formed on basis of these, grade of difference between acoustic signal and standards is determined, on basis of aforementioned grade decision is taken whether acoustic signal belongs to person, whose values of statistical characteristics were used during forming of standards.

EFFECT: higher resistance to interference, higher efficiency, higher trustworthiness.

6 dwg

FIELD: audio signal compression technologies.

SUBSTANCE: method includes first determining of whether current audio signal contains information in form of speech or noise, while second determining stage is performed, whether signal contains non-speech data, being important for sensing by listener, and selective cancellation of result of first determination is performed, appropriate for noise, in response to result of second determination, appropriate for non-speech data, being important for sensing by listener.

EFFECT: higher efficiency.

3 cl, 13 dwg

The invention relates to the field of speech analysis, in particular to restrict unauthorized access to the material or information resources on the basis of biometric information"

The invention relates to speech recognition

FIELD: audio signal compression technologies.

SUBSTANCE: method includes first determining of whether current audio signal contains information in form of speech or noise, while second determining stage is performed, whether signal contains non-speech data, being important for sensing by listener, and selective cancellation of result of first determination is performed, appropriate for noise, in response to result of second determination, appropriate for non-speech data, being important for sensing by listener.

EFFECT: higher efficiency.

3 cl, 13 dwg

Up!