RU2278424C1 - Device for measuring maximal legibility of speech - Google Patents

Abstract

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

Description

The field of technology.

The invention relates to speech recognition, specifically to measuring the maximum possible speech intelligibility in extremely adverse conditions, in noise, at low signal-to-noise ratios, and is intended mainly to determine the degree of protection of objects from leakage of speech information through several leakage channels simultaneously, t .e. in real conditions.

The prior art.

Various devices for measuring speech intelligibility are known. Expert listening is popular at the place of receiving acoustic signals composed of specially selected phrases, words or syllables (programs) that are emitted at the place of speech generation, with subsequent processing of the results to determine intelligibility (P.M. Polkovsky, A.D. Tkachenko. feedback paths. - M .: Communication, 1969; GOST 71-53-85 "Measurement of intelligibility by articulation or tone method"; GOST B 20775-75. "Voice transmission over communication paths equipped with security equipment. Speech intelligibility requirements. Method of measurement"). In these works, similar measuring devices are described, the apparatus used and the methods of mathematical processing used. The devices are technically simple, but the measurements are not sufficiently reproducible and accurate in conditions of weak acoustic signals, since it depends on the intelligence of experts, and most importantly, they are laborious and require the participation of a large number of experts.

More objective is the method and devices based on the sequential acoustic radiation of individual tonal components of the audible spectrum at the place of transmission and the determination by any experts of the threshold of their audibility at the place of reception, and in fact at the place of reception there can only be microphones connected through amplifiers to the head phones of experts (N. B. Pokrovsky. Calculation and measurement of speech intelligibility. - M.: Communication, 1962). The level of acoustic radiation of each component is increased to the threshold of hearing by its expert. The fixed radiation levels are mathematically processed taking into account the human hearing curve and thus determine the minimum level of the speech signal to obtain a given speech intelligibility at the reception site. Naturally, the obtained information can also be used to determine the maximum level of the acoustic signal at which speech will be guaranteed to be illegible at a given place of reception, i.e. a predetermined level of protection against leakage of acoustic information is provided. However, this method is laborious, since it also requires the participation of many experts and time. In addition, intelligibility is assessed not at the real level of acoustic signals, but at those that correspond to the threshold of audibility in a given place. This is often unacceptable, since, as you know, spurious channels of transmission of acoustic information (walls, floors, windows, furniture, doors, etc.) usually unevenly weaken the frequency components of speech and the degree of protection of information from leakage, calculated from the results of the described measurements, can not match the real ones.

Currently, the most common hardware methods based on the tonal method of speech intelligibility (for example, P.M. Polkovsky, A.D. Tkachenko. Electro-acoustic paths with feedback. - M .: Communication, 1969, GOST B 20775-75, Utility model No. 27259). They provide for the measurement of intelligibility at any signal to noise ratio exclusively by technical means, without the use of expert listening. The most modern device according to the patent for utility model No. 27259, publ. 01/10/2003. It is closest to the claimed device for most of the essential features and is therefore selected as a prototype.

The following is a detailed description of the prototype device.

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

The device consists of a series-connected generator 1 of a test signal, a loudspeaker 2, located at the actual location of the source of sound signals, a microphone 3, located at the listening position, an N-band meter 4 of signal-to-noise ratios at each of the N frequencies of the test signals and a level 5 calculator speech intelligibility.

The device operates as follows. After placing the speaker 2 and the microphone 3 in the indicated places, the device is turned on. Generator 1 generates a test acoustic signal of a given level, consisting of a sequence of N frequencies spaced apart from the middle frequencies of N bands into which the audible speech spectrum is divided. Loudspeaker 2 emits the indicated signals. Microphone 3 converts the signals (received at the indicated radiations and in the pauses between them) into electric ones. During a pause, only the noise signal in this band is received and measured by the same receiver. According to the measurement results, the power values of the received signal of this tonal frequency are determined based on the independent linear addition of the signal and noise powers, and the signal-to-noise ratios at this tonal frequency are determined. The procedure is then repeated at the next tonal frequency, then at the next until the signal-to-noise ratios at all tonal frequencies of the test signal are determined. This occurs in an N-band 4 signal-to-noise ratio meter. Thus, at the input of the calculator 5 of the intelligibility level, the measured signal-to-noise ratios for all N radiated frequencies are received in turn. The intelligibility level calculator 5 for the measured signal-to-noise ratios calculates speech intelligibility. The calculation is based on known relationships and theories, the essence of which is that first, the intelligibility factors at each frequency of the test signal are determined from the above measured signal-to-noise ratios. For certain signal-to-noise ratios at all tonal frequencies of the test signal, the speech intelligibility factor is calculated against the background of real noise - based on well-known techniques (see, for example, article AAKolesnikova, V.F.Komarovich, V.K.Zheleznyak. Correlation theory of intelligibility speech in the journal "Questions of Radio Electronics", series "General Questions of Radio Electronics", 1995, issue 1, pp. 3-9; NB Pokrovsky. Calculation and measurement of speech intelligibility. - M .: Communication, 1962).

They determine the average (arithmetic mean) value of the intelligibility factor. Then, according to a certain arithmetic mean of the intelligibility factor and the known curve of the dependence of verbal intelligibility on the value of the intelligibility coefficient, verbal intelligibility of speech is determined (for more details see, for example, the article by V.K.Zheleznyak et al. "Special equipment" 2000, No. 4, pp. 39-45).

In other words, the essence of the device, like most other devices for measuring intelligibility, consists in sequential radiation at the transmission point of test N tonal acoustic oscillations in the speech frequency range, acoustic reception of the received signals at a given reception point, N-band measurement of signal-to-noise ratios at all N frequencies of the test signal and the subsequent calculation of the speech intelligibility coefficient from them - by known methods.

The described devices are widely used for measuring speech intelligibility, i.e. to determine the quality of listening. However, at present, the fight against eavesdropping, i.e. reduction of intelligibility in places of possible information leakage, prevention of unauthorized listening. In this case (in contrast to the task of determining the quality of listening), the signal-to-noise ratio is small and it is difficult for an attacker to parse the speech signal. As you know, in such cases, not only acoustic, but also all other possible channels of information leakage are used for eavesdropping, capturing and processing, in particular, magnetic fields generated by sound amplifying equipment, and vibro-acoustic vibrations associated with its operation. The signals from these channels may in some cases be more intelligible than acoustic signals. In such cases, the use of the described device to control speech intelligibility is not acceptable, it does not reflect the possible level of speech intelligibility.

Naturally, the above-described acoustic intelligibility measuring device can be modified and adapted to measure speech intelligibility through other possible information leakage channels, the most important of which are magnetic, electrical and vibro-acoustic channels. Then it will be possible to evaluate the level of maximum intelligibility as the largest of all the channels listed. However, these devices cannot measure the real level of protection of speech information from unauthorized listening, since when using all possible leak channels to listen to recordings of signals, it is possible to further jointly process these records to extract speech information (see, for example, A.G. Okhonsky, A .A.Aliseev et al. Immunity of information radio control systems: A Training Manual. Publishing House of MGAP "Book World", 1993, which describes the scheme of the optimal discriminator of two deterministic signals, m the most plausible estimation of the signal parameter with the initial phase, optimal signal processing for useful signal models with a random initial phase in the form of a quadrature signal processing, which can provide a intelligibility level greater than the maximum in any of the individual channels. As is known, each channel of information leakage is frequency-dependent. Accordingly, the frequency dependences of the signal-to-noise ratios of the channels are different. The signal-to-noise ratios will be different in the same frequency band over different leakage channels. Based on this, it is hypothetically possible the most sensitive, most rational increase in speech intelligibility, providing for the separation of the signals obtained for analysis of each channel into several frequency bands and the enumeration of all possible combinations of channel frequency band signals among themselves to obtain the most intelligibility. Implementing such an eavesdropping device is complex, expensive, but nonetheless possible.

To assess the protection of sound reinforcement equipment with such a multi-channel eavesdropping method (which, to our knowledge, has not yet been implemented anywhere in hardware, but is quite feasible), a technique for measuring maximum intelligibility is necessary, taking into account its extreme capabilities. We called it the measurement of "maximum intelligibility."

The known device is a prototype of such a measurement can not provide.

The objective of the invention.

The objective of the present invention is to increase the reliability of measuring speech intelligibility with multiple channels of speech information leakage.

The solution to the problem is the essence of the invention.

The problem is solved in that in the known device for measuring speech intelligibility, containing a series-connected acoustic test signal generator in the form of a sequence of N frequencies distributed over the middle frequencies of the bands into which the audible spectrum is divided, with pauses between frequencies, an emitter, an acoustic signal receiver, N -band signal-to-noise ratio meter and intelligibility calculator, the generator and the N-band meter having synchronously controlled switching clock ate bandwidths substantial changes and additions, namely:

- parallel to the receiver of the acoustic signal installed To the receivers of other types of signals formed by the radiation of the acoustic test signal, for example, magnetic, electrical, vibroacoustic,

- at the receiving point, K receivers of other types of signals formed by an acoustic test signal, for example, electric, magnetic, vibroacoustic signals, are additionally placed,

- each frequency of the N-frequency test signal and the pause are repeated cyclically K + 1 times.

- the outputs of the acoustic receiver and K receivers of other signals are connected to the input of the N-band signal-to-noise ratio meter through the receiver switch, synchronized with the beginning of each of K + 1 repetition cycles,

- the output of the N-frequency signal-to-noise ratio meter is connected to the input of the intelligibility calculator through a selection device in each cycle of the largest values of the signal-to-noise ratios at each frequency of the test signal,

- in this case, the clock generator has additional outputs for organizing the generator cycles, controlling the receiver switch and the selection device.

Disclosure of the invention.

The essence of the invention is to provide intelligibility measurements at the largest of all types of received signals, signal-to-noise ratios in each frequency band. With unauthorized eavesdropping, the conditions for obtaining maximum intelligibility even theoretically cannot be the best. Thus, we determine the maximum speech intelligibility.

An example of a simplified structural diagram of the device is presented in figure 1, where the following notation is adopted: 1 - generator of an N-frequency test signal, 2 - acoustic emitter (loudspeaker), 3 - receiver of acoustic signals (microphone), 4 - N-band signal ratio meter / noise, 5 - calculator of the level of speech intelligibility, 6 - receiver of magnetic signals, 7 - receiver of electrical signals, 8 - receiver of vibro-acoustic signals, 9 - switch of receivers, 10 - device for choosing the highest signal-to-noise ratio, 11 - clock sync lower. In the example, the number of other types of receivers is K = 3 (magnetic, electric and vibro-acoustic).

The device consists of a series-connected generator 1 of an N-frequency test signal, a loudspeaker 2 located at the actual location of the source of speech signals, microphone 3, receivers 6, 7, 8 of magnetic, electrical and vibro-acoustic signals located at the listening position, switch 9 of the receivers, An N-band meter 4 signal-to-noise ratios, a device 10 for selecting the largest signal-to-noise ratio, and a speech intelligibility level calculator 5.

The operation of the device is synchronized by such a synchronizer 11.

First, place the source of the acoustic test signal (loudspeaker 2) at the point of placement of the source of the speech signal, and the receivers of acoustic 3, magnetic 6, electric 7 and vibro-acoustic 8 signals at the point of reception of voice information.

The generator 1 generates, the emitter 2 emits a test acoustic signal of a given level, which is, most often, a sequence of tonal frequencies distributed over the middle frequencies N of the bands into which the audible spectrum is divided, with pauses between the frequencies. In this case, the radiation and pause at each frequency are cyclically repeated K + 1 = 4 times.

Switching in each cycle with the switch 9, the receivers 2, 6, 7, 8 are taken one by one, they also receive an N-band meter 4, the received acoustic, magnetic, electric and vibro-acoustic signals are measured at the receiving point during the radiation and in the pauses between them. In this case, the received signals are divided by N narrow-band filters into frequencies corresponding to the frequencies of the test signal. Narrow-band filters are switched by the synchronizer 11 simultaneously with a change in the frequency of generation of the test signal. Since the levels of the received signals are measured at the output of each receiver (in our example, acoustic, magnetic, electric, and vibroacoustic), four signal-to-noise ratios are determined at each frequency of the test signal. Then, the selection device 9 selects at each frequency of the test signal (i.e., each band of the audible spectrum) the largest signal-to-noise ratio. The operation of the switch 9 of the receivers and the device 10 for selecting the largest signal-to-noise ratio is synchronized by the clock synchronizer 11. Thus, the input of the intelligibility level calculator 5 alternately receives the highest signal-to-noise ratios for all N radiated frequencies. The intelligibility level calculator 5 for the measured signal-to-noise ratios calculates speech intelligibility. The calculation is based on known relationships and theories, the essence of which is that first, the intelligibility factors at each frequency of the test signal are determined from the above measured signal-to-noise ratios. They determine the average (arithmetic mean) value of the intelligibility factor. Then, according to a certain arithmetic mean of the intelligibility factor and the known curve of the dependence of verbal intelligibility on the value of intelligibility coefficient, the maximum verbal intelligibility of speech is determined (for more details see, for example, in the article by V.K. Zheleznyak et al. "Special equipment", 2000, No. 4, pp. 39-45).

The main differences of the claimed device from the prototype:

- parallel to the receiver of the acoustic signal installed To the receivers of other types of signals formed by acoustic radiation, for example, magnetic, electrical, vibro-acoustic,

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

- the outputs of the acoustic receiver and K receivers of other signals are connected to the input of the N-band signal-to-noise ratio meter through a receiver switch switched by a synchronizer at the beginning of each of the K + 1 cycles,

- the output of the N-frequency signal-to-noise ratio meter is connected to the input of the intelligibility calculator through a selection device in each cycle (i.e., at each frequency of the test signal) of the highest signal-to-noise ratio,

- in this case, the clock generator has additional outputs for organizing the generator cycles, controlling the receiver switch and the selection device.

Such a device is not known from available sources of information. The totality of these differences is a creative achievement by the inventor, it is not obvious to a specialist.

Industrial applicability.

The claimed device is easily feasible, both schematically and programmatically. Scheme - means well mastered by the industry, which follows from the above description of an example implementation.

In particular, the receiver of acoustic signals can be a microphone, magnetic signals, a magnetic antenna (see, for example, RF patent 2152624 "Measuring the intensity of the magnetic component of an alternating electromagnetic field", electrical signals - an electric antenna (see, for example, RF patent 2152623 component of an alternating electromagnetic field ", vibroacoustic signals - a piezoelectric transducer (see, for example, I.K. Kolesnikova and others. Fundamentals of hydroacoustics and hydroacoustic stations.- L .: Shipbuilding, 1970) The choice of the highest signal-to-noise value can be performed, for example, on the basis of a circuit or with memory circuits.

The remaining components of the device — generators of a sequence of N test signals, emitters, an N-band signal-to-noise ratio meter, a intelligibility level calculator — were repeatedly implemented in well-known intelligibility measuring devices (see, for example, certificate 27259 for a utility model, published on January 10, 2003, Bull. No. 1 for 2003)

The claimed device was implemented at the Federal State Unitary Enterprise Informakustika, successfully passed metrological and certification tests and is already used in measurements of maximum speech intelligibility. However, no practical difficulties associated with the use of the invention were encountered.

The real advantages over the prototype are to increase the reliability of measuring the level of speech intelligibility. The measurements taken guarantee a truly potentially maximum intelligibility, taking into account the use by the listening person of all possible channels for the leakage of speech information. In particular, the maximum intelligibility measured during the tests when using only one magnetic channel in addition to the acoustic channel turned out to be 5 dB higher than that measured by the prototype method. Such an error of known measuring devices is unacceptable.

The achieved reliability of the definition of speech intelligibility makes it possible to establish with guarantee the absence of information leakage channels, since intelligibility is determined based on the use of all possible leakage channels and the most rational processing of their signals.

Thus, in our opinion, the claimed technical solution meets all the criteria for inventions - it is new, non-obvious and industrially applicable.

Claims (1)

A device for measuring maximum speech intelligibility, comprising a series-connected acoustic test signal generator in the form of a sequence of N frequencies distributed over the middle frequencies of the bands into which the audible spectrum is divided, with pauses between frequencies, emitter, acoustic signal receiver, N-band signal-to-noise ratio meter and a intelligibility calculator, the generator and the N-band meter having synchronously controlled by the clock synchronizer bandwidth switches, distinguishing In that parallel to the acoustic signal receiver, K receivers of other types of signals formed by the radiation of the acoustic test signal, for example, magnetic, electric, vibroacoustic, are installed, each frequency of the sequence of the N-frequency test signal and the pause are repeated cyclically K + 1 times, the outputs of the acoustic receiver and To the receivers of other signals are connected to the input of the N-band signal-to-noise ratio meter through the receiver switch, synchronized with the beginning of each of the K + 1 cycles of test frequency, the output of the N-frequency signal-to-noise ratio meter is connected to the input of the intelligibility computer through the selection device in each cycle of the highest values of signal-to-noise ratios, while the clock generator has additional outputs for organizing K + 1 generator cycles and synchronous control of the receiver switch and a device of choice.

Images