DE1537596B1 - Method and circuit arrangement for improving the intelligibility of frequency-distorted speech - Google Patents

Description

The invention relates to a method and: a circuit arrangement to improve the intelligibility of speech produced as a result of an increase in the Spokesman because of burdensome pressure and / or because of a foreign atmosphere acting accordingly is distorted.

The invention finds advantageous application in particular in the transmission of speech by divers who operate at a considerable depth below the Working on the surface of the water and having to speak in a room, the air you breathe and a Contains helium gas mixture. The helium gas mixture serves to reduce the accumulation of nitrogen in the diver's bloodstream, which prevents the otherwise occurring The anesthetic effect is avoided or at least reduced.

In normal audible speech, every word has a frequency spectrum, that is familiar and can be understood by the human ear. While talking under the conditions of a helium mixture: atmosphere and an abnormal pressure, z. B. 60 m below the water surface, however, each of these abnormal conditions is effective on the human speech mechanism, so that the speech frequency spectrum in distorted in a nonlinear manner and incomprehensible or essentially incomprehensible will.

The German patent 510 099 describes a method for transmission of a waveband over a conduction path that is used for certain, especially higher Frequencies has very high impedance, lowering all frequencies of the band became known on frequencies that are in the favorable transmission range of the line, in which, for example, by means of thermionic frequency modulators a certain Frequency of the band to be transmitted is subtracted, so that the width of the degraded Tape is the same as that of the original tape intended for transfer remain. Only the effective parts of the frequency band are reduced but the remaining parts are completely suppressed.

Furthermore, through the German patent specification 962 268, a system for the transmission of messages with a wide frequency range within a narrow transmission band through a selection of narrow sections from the whole News band became known, which is characterized in that the frequency position of the sections to be selected is determined in advance, that between selected sections Frequency band gaps remain that are not transmitted, that all selected Sections can be shifted with the help of modulators in such a way that they form a contiguous Form tape and that means for marking the original position of the cutouts and provided for transmitting the compressed tape and the identification signals are.

Compared to this known prior art, the invention is the Object is based on a method and a circuit arrangement for converting Speech with an essentially incomprehensible spectrum as a result of the above non-linear speech frequency distortion in speech with an intelligible speech spectrum to accomplish.

This object is achieved according to the invention in that from the original speech spectrum at least two separate frequency ranges which are a substantial part of the original spoken message contain that these areas, with the exception of the area with the lowest frequencies screened out depending on the pressure and / or the foreign atmosphere be shifted to lower frequencies and closer to each other and that the Output signals of these shifted areas are mixed.

Preferably, the screened frequency ranges are first after higher frequencies and then shifted to lower frequencies, the Shifting each range to lower frequencies from the pressure and / or the Foreign atmosphere depends and takes place in such a way that by the relative area shifts the areas have a smaller frequency spacing from one another than the original screened areas.

Advantageous developments of the method result from the features of claims 3 and 4. To solve the underlying of the invention, given above The task is also a circuit arrangement for carrying out the described method, which is characterized according to the invention that for frequency shifting for a modulator and oscillator device for displacement of each of the screened-out areas the areas for higher frequencies and a demodulator and oscillator device for shifting the ranges towards lower frequencies and filters for elimination unwanted frequencies are provided.

Advantageous refinements of this circuit arrangement result from the features of claims 6 and 7.

Briefly stated, the invention includes a method and circuit arrangement to select separate frequency ranges of the original speech spectrum and shifting the ranges to lower frequencies by different degrees in Depending on the environmental conditions described above, to get an improved To create speech intelligibility.

Below is a preferred embodiment of the invention described with reference to the drawing. It shows: F i g. 1 is a block diagram of the essential parts of the method and the circuit arrangement according to the System embodying the invention, FIG. Figure 2 is a block diagram of a detail of one modified system according to F i g.1, F i g. 3 is a diagram of two selected areas before and after the frequency shift; F i g. 4 a, 4 b and 4 c a detailed one Circuit of the arrangement according to the invention.

F i g. 1 illustrates a block diagram of an electrical system 10, which the method and the circuit arrangement according to the invention, for example illustrated.

In the system 10, the distorted speech of the diver is converted into electrical signals by a converter 12 which is connected to an input amplifier 16 via the line 14. The amplified signal is fed through lines 18, 20 and 22 to the bandpass filters 24 and 26, which screen out separate frequency ranges of the original speech spectrum. Thus, for example (Fig. 3) the areas B and C can be obtained by the filters 24 and 26, respectively, the area B the frequencies from 850 to 1700 Hz and the area C the frequencies -from 2400 to 5000 Hz, so that the areas are originally separated by a distance of 700 Hz.

According to the invention, the selected areas are then unequal Reduced amounts, d. H. shifted towards lower frequencies to the original modify distorted speech spectrum and relate the areas to the normal language spectrum to: bring. To better understand the problem as well of the method and the circuit arrangement for its solution, it should be noted that the language of the diver under abnormal pressure conditions, for example in 61 m depth below sea level, and the foreign gas environment, for example a Mixture of 3 ° / ° oxygen, 80 ° / ° helium and 171 / d nitrogen, becomes incomprehensible and like the incomprehensible voice of Walt Disney's character Donald Duck listens. According to the invention it has been found that by selecting separate Frequency ranges of the original distorted speech and frequency reduction of the selected areas and moving them towards each other, namely in Dependence on the percentage of helium in the surrounding atmosphere and on pressure on the diver, a significant improvement in the language spectrum can be obtained and the language becomes intelligible.

Accordingly, there are in the system 10 per se. known means of implementation the frequency conversion described above is provided. In the embodiment of the system 10 (FIG. 1) have separate modulation and filtering devices connected to the output of each of the filters 24 and 26.

The output of the filter 24 is through a line 28 with a modulator 30, which is also connected to the output of an oscillator 32 via a line 34. The output of the modulator 30 is available via a line 36 with a high-pass filter 38 in connection to exclude unwanted frequencies. The output of the filter 38 is led through a line 40 to a demodulator 42, which via a line 46 is at the output of a tunable oscillator 44. The output of the demodulator 42 is connected by a line 48 to a low-pass filter 50 in order to avoid undesired Exclude frequencies.

In the same way, the area C, i. H. the output signal the filter 26, scaled down by providing a similar group; which from a modulator 52 and an associated oscillator 58, a high-pass filter 62, a demodulator 66 and an associated oscillator 70 and a low-pass filter 74 consists. The elements are connected in the same way as for those with the Filter 24 connected group described. That is, the modulator 52 is on the Lines 54 and 56, respectively, are connected to the filter 26 and the oscillator 58. The exit the modulator 52 goes over the line 60 to the filter 62, then over the line 64 to demodulator 66; which is connected to the oscillator 70 by the line 68 is, and finally via the line 72 to the filter 74.

The oscillators 32 and 58 serve to translate the areas B and C. For this purpose, two oscillators with 8000 Hz each were used. the Frequencies of the oscillators 44 and 70 are selected; that the resulting Displacement, as shown in FIG. 3 indicated, a. Shift of areas B and C to lower frequencies as well as a movement of these areas towards each other where the displacement of each area and the ratio of the displacement of areas, of conditions. of the helium mixture and the surrounding pressure among which the diver is speaking:.

F i.g. 3 shows the results obtained by the above-described System.- The band pass filters 24 and 26 seven. an area B with a frequency range from 850 to 1700 Hz or a range C with a frequency range from 2400 to 5000 Hz, the areas B and C by a frequency spacing separated from 700 Hz. The areas B. and. C are therefore selected from one another .divided parts. of the original distorted speech spectrum.

The areas B and C can be reduced and shifted towards each other by the oscillators 32 and 58 of 8000 Hz each and the oscillators 44 and .. 70- of 8600 or 9000 Hz, in order to move the area B 'with a frequency range of 250 to 1100 Hz as well as the converted range C with a frequency range of 1400 to 4000 Hz. The modulator 30, the oscillator 32 and the filter 38 translate the range B around 8000 Hz into a range (not shown) from 8850 to 970014z, while the demodulator 42, the oscillator 44 and the filter 50 convert the range around 8600 Hz into one Reduce range B 'from 250 to 1100 Hz (FIG. 3). In the same way, the modulator 52, the oscillator 58 and the filter 62 translate the range .C around 8000 Hz into a range (not shown) from 10,400 to 13,000 Hz, while the demodulator 66; the oscillator 70 and the filter 74 reduce the range around 9000 Hz in the range C from 1400 to 4000 Hz (FIG. 3). The converted areas B 'and C therefore have a lower frequency and are closer together (that is, separated only by a frequency spacing of 300 Hz) than the original distorted areas B and C.

As in Fig. 1, the output signals of filters 50 and 74 are fed through lines 76 and 80 or 78 and 80 to a mixer 82, the output of which is connected through line 84 to a suitable receiver, for example headphones 86.

It should be noted that the degree of nonlinear distortion the original language with the percentage of helium in the environment in which the speaker is located, changes and also depends to some extent on the pressure, to which the speaker is exposed. While improving speech intelligibility achieved by the method and the circuit arrangement as described above being a conversion of the original selected areas into lower ones Frequencies and towards each other are made further improvements through special Select the oscillators 44 and 70 obtained from the specific conditions of the helium percentage and the pressure depend for a maximum improvement to achieve speech intelligibility.

The oscillators 44 and 70 may be non-controllable oscillators selected for an average of the environmental conditions, and yet the system provides improved intelligible speech under various environmental conditions. However, according to the invention, manually adjustable oscillators 44 and 70 can also be provided. An advantage of this arrangement is that the system can be adapted to the changes in the environmental conditions to which the speaker is subjected and to maximum speech intelligibility. A second advantage is that by listening to the corrected speech and adjusting the oscillators 44 and 70 while listening, the system operator can achieve maximum intelligibility of the speech.

The above-described and in FIG. 1 system shown was with satisfactory results under various helium mixtures and pressures tested corresponding to a depth of 61 to 91.5 m. Under these conditions a Increase in speech intelligibility from 20 to 90%.

It is also possible to increase the intelligibility by increasing the number the originally selected language spectrum areas and conversion of the same to further increase the manner described above. The practical limit The number of areas selected seems to be a matter of cost in the system used parts and the need to be some distance between adhere to neighboring areas in order to avoid mutual modulation.

It is also possible, as shown in FIG. 3 shown to screen out a lower spectrum range without converting the same. In Fig. 2, a bandpass filter 100 is therefore added, which is passed through lines 102 and 104 between the circuits shown in FIG. Lines 20 and 76 shown in FIG. 1 are connected in order to feed the mixer stage 82 directly. The filter 100 provides a range below the spectrum of the range B 'and at a certain frequency spacing therefrom.

With regard to the method and the circuit arrangement according to the invention it should be noted that deep-sea divers are currently using key words to tell each other what the extent of the information to be transmitted is severely limited. When the working depth and / or the ratio of helium in the breath increases, even one becomes strong Limited communication difficult and often impossible. When an unforeseen Information becomes necessary, d. H. information describing a found State that cannot be communicated using previously agreed keywords, the language very often becomes incomprehensible and, at best, requires frequent repetition as well as a person who is extremely experienced in listening to this type of speech.

With the method and the circuit arrangement according to the invention can but also transmit an unforeseen speaking about any object and be received in an understandable way. The importance of the invention is that for the first time a coherent unforeseen voice message between Sea diving among each other and between the divers and the escort ship is possible is. Since a voice connection in deep-sea diving operations, such as. B. Investigation, Exploration, rescue and recovery, is important and given the equipment for a diver for long-term work in deep sea conditions for several years is available, the speech connection that can be understood according to the invention creates the essential one Supplement so that deep-sea divers can now work safely and successfully.

4a, 4b and 4c show in detail the parts of an example Embodiment of the circuit according to the invention, which has been tested and excellent Results on improving the intelligibility of speech showed which is distorted by ambient conditions of a helium mixture and increased pressure.

The values of the in FIGS. 4a, 4b and 4c are given in the following table: Individual values for the in the Fi g. 4a, 4b and 4c parts shown R1 18 K R52 12 K .R2 10-33 R53 1.2 K R3 12 K R54 100 R4 1.2 K R55 390 R5 1.5 K R56 3.3 K R6 100 R57 56 K R7 390 R58 3.9 K R8 56 K R59 18 K R9 3.3 K R60 10-33 R10 3; 9 K R61 150 K R11 56 K R62 12 K R12 750 R63 1.2 K R13 560 R64 100 R15 1.8 K R65 390 R16 2.7 K R66 56 K R17 22 K R67 3.3 K R18 470 K R68 3; 9 K k19 62 R691 10 K R20 2; 7-K R70 3.3 K R21 1.2 K R71 56 K. R22 27 R72 18 K R23 12 K R73 10-33 R24 6.8 K R74 12 K R25 6.8 K R75 1.2 K R26 1 K R76 100 R27 1 K R77 390 R28 2.2 K R78 150 K R29 1 K R79 10 R30 2.2 K R80 2.2 K R31 2.2 K R81 1 K R32 1 K - R82 2.2 K R33 2.2 K R83 2.2 K R34 1 K R84 1 K R35 33 K R85 2.2 K R36 10 K R86 1 K R37 4; 7 K R87 33 K R38 2.2 K R88 10 K R39 1 K R89 232 K R40 2.2 K R90 1 K R41 2.2 K R91 2.2 K R42 1 K R92 2.2 K R43 2.2 K R93 1 K R44 6.8 K R94 2.2 K R45 1 K R95 4.7 K R46 33 K R96 10 K R47 4.7 K R97 33 K R48 10 K R98 1 K R49 150 K R99 6.8 K R50 18 K R100 56 K R51 10-33 R 101 750 continuation Individual values for the in the F i g. 4 a, 4 b and 4 c parts shown R102 750 C22 4.7 35V R103 560 C23 no value R104 2.1 K C24 4.7 35 V. R 105 1.8 K C25, 33 R 106 62 C26, 33 R 107 2; 7 K C27, 27 R108 22 K C28; 062 R109 470 K C29, 27 R110 1.2 K C30, 062 R111 27 C31 82 50 V R112 12 K C32 82 50 V R113 680 C33; 33 R114 1 K C34, 01 R115 330 C351 0-0.1 R 116 100 C36} R117 22 K C37, 33 R118 33 C38, 01 R119 200 C391 0-0.1 R120 1 C40 R 121 470 C41, 1 R122 1 K C42, 1 R123 1 K C43 82 50 V R 124 1 K C44 82 50 V R 125 1 K C45 4; 7 R 126 560 C46 4.7 R127 560 C47 4.7 35V R128 560 C48 no value R200 220 C49 4.7 35 V. C1, 047 35V Q1 300C022-1 C2; 047 35 V Q2 2N526 C3, 047 35V Q3 300C022-1 C4, 047 35 V Q4 2N526 C5 4; 7 10V Q5 300C022-1 C6 4.7 10V Q6 2N526 C7 4.7 10V Q7 300C022-1 C8 4; 7 10V Q8 2N526 C9 15 20 V Q9 2N465 C 10 15 20 V Q10 2N539A C il 15 20 VQ 11 300C022-1 C12 15 20 V Q12 300C022-1 C13 82 50 VQ 13 300C022-1 C14 82 50 V Q14 300C022-1 C15 82 50 V Q15 2N465 C16 82 50 VQ 16 2N539A C17 4; 7 30V Q.17 2N43A C18 4.7 30 VQ 18 2N43A c19 4; 7 30 VQ 19 2N539A C20 4.7 30V Q20 2N539A C21 4.7 30V

Claims (7)

Claims: 1. Method for improving the intelligibility of speech which is distorted as a result of an increase in the pressure on the speaker and / or because of a correspondingly foreign atmosphere; characterized in that at least two separate frequency ranges (B and C) are sifted out from the original speech spectrum and contain a substantial part of the original spoken message / or the foreign atmosphere are shifted to lower frequencies and closer to one another and that the output signals of these shifted areas (B 'and C ) are mixed (FIG. 3). 2. The method according to claim 1, characterized in that the sieved frequency ranges are first shifted to higher frequencies and then to lower frequencies, the shift of each area to lower frequencies depends on the pressure and / or the foreign atmosphere and is carried out by the relative area shifts the areas (B 'and C ) have a smaller frequency spacing from one another than the areas originally screened out (B and C; FIG. 3). 3. The method according to claims 1 and 2, characterized in that the width of the screened areas (B and C) and the relative shift to lower frequencies are chosen so that a minimum distance between the shifted areas (B 'and to avoid cross-modulation C ) remains (Fig. 3). 4. The method according to claims 1 and 2, characterized characterized in that the degree of shifting of the screened areas to lower Frequencies are regulated by hand for maximum intelligibility. 5. Circuit arrangement for performing the method according to claims 1 to 4, characterized in that for frequency shifting for each of the screened out Areas a modulator and oscillator device (30, 32 or 52, 58) for displacement the areas for higher frequencies and a demodulator and oscillator device (42, 44 or 66, 70) to shift the ranges towards lower frequencies and filters (38; 50 or 62, 74) are provided for eliminating undesired frequencies are (Fig. 1). 6. Circuit arrangement according to claim 5, characterized in that that the oscillator device (44 or 66) connected to the demodulator (42 or 66) 70) can be controlled manually (FIG. 1). 7. Circuit arrangement according to claim 5, characterized characterized in that the filters and the frequency shifting devices are so sized are that the screened areas (B and C) after the shift to lower Frequencies a sufficient frequency spacing to avoid cross modulation keep (Fig. 3).