GB2131659A

GB2131659A - Sound synthesizer

Info

Publication number: GB2131659A
Application number: GB8318893A
Authority: GB
Inventors: Fumitada Itakura; Noburu Sugamura
Original assignee: Nippon Telegraph and Telephone Corp
Current assignee: Nippon Telegraph and Telephone Corp
Priority date: 1979-10-03
Filing date: 1980-09-24
Publication date: 1984-06-20
Also published as: GB2059726B; GB8318893D0; DE3037276A1; GB2131659B; SE8006850L; DE3050742C2; NL8005449A; NL189320C; FR2466826B1; SE444730B; DE3037276C2; FR2466826A1; GB2059726A; NL189320B

Abstract

A sound synthesizer in which pulses of a period indicated by a fundamental period parameter are produced by a fundamental period sound source, the output from the fundamental period sound source or the output from a noise source is selected depending on whether a sound to be synthesized is a voiced or unvoiced sound, and the selected pulses are applied to a sound synthesis filter section 16 to synthesize the sound. The sound synthesis filter section is composed of second-order filter means (57-61, 65-69) which serves as a second- order filter having the zero on a unit circle in a complex plane, means for cascade-operating second-order filter means of different coefficients, and feedback means 41, 42 for feeding back the output from the synthesis filter section 16 to the input side thereof through two kinds of such cascade- operating means, and coefficients of the second-order filters are controlled by control parameters which are derived from line spectrum pair parameters (LSP) the origination and conversion of which are discussed in the specification. <IMAGE>

Description

1

SPECIFICATION

Sound synthesizer The present invention relates to a method of sou nd synthesizing and to a sound synthesizer with wh ich it is possible to reconstruct a sound of substantially the same quality as an original sound from its features transmitted or stored in a memory in a small amount of information.

For example, in the case of reconstructing speech from feature parameters of original speech, according to the priaradthe output of a pulse generator simulating thevibration of the vocal cord and the output of a noise generator simulating turbulence are changed over armixed together depending on whetherthe speech isvoiced or unvoiced and the resulting output is amplitude-modulated in accordancewith the speech amplitudeto produce an excitation sourcesignal which is applied to afilter simulating the resonance characteristics of the vocal tractto obtain synthesized speech. Asynthesis system using partial. auto correlation (PARCOR) coefficients and a formatsynthesis system are examples of such speech synthesis system employing the featu re parameters. The former is set forth, for example, in J. D_ Markel et aL, - Linear Prediction of Speech-, pages 92-128, Springer-Verlag, 1976, in which the partial auto correlation coefficients orthe so-called PARCOR coefficients of a speech waveform are used as thefeature parameters. If the absolute values of the PARCOR coefficients are all smallerthan unity, the speech synthesizing filter is stable. The PARCOR coefficients may be relatively small in the amount of information for speech synthesis andthe automatic extraction of the coefficients is relatively easy, butthe individual parameters differwidely in the spectral sensitivity. Accordingly, if all the parameters are quantized using the same number of bits, spectral distortions caused by quantization errors forthe respective parameters largely differfrom each other. Furtherthe PARCOR coefflcientsare poorin-their interpolation characteristics and, bythe interpolation ofthe parameters, there are produced noises, resulting in an indistinct speech. Especially ata low bit rate, the speech quality is deteriorated by.the spectral distortion and no satisfactory synthesized speech quality is obtainable. In addition, sincethe PARCOR coefficients do not directly correspondto spectral properties such as formant frequencies, and hencethe PARCOR coefficients are not suitable for speech synthesis by rule.

The formant synthesis system is disclosedJor example, in J.. L. Flanagan,--SpeechAnalysis, Synth- esis and Perception% pages 339-347, Springer-Verlag, 1972. This system is one which synthesizes speech using the formantf requencies and their intensity as parameters and which is advantageous in thatthe amount of information forthe parameters may be small and in thatthe correspondance of the parameters to spectral quantities is easyto obtain. Fortheextraction of the formantfrequency and the intensity thereof, however, it is necessaryto make use of general dynamic characteristics and statistical properties of the parameters, and complete automa- GB 2 131 659 A 1 tic extraction of theformantfrequency and the intensity thereof is difficult. Accordingly, it is difficult to automatically obtain synthesized speech of high quality and it is likelyto markedly degradethe quality of the synthesized speech by an error in the extraction of the parameters.

It is an object of the present invention to provide a sound synthesizer which is able to synthesize a sound of high quality using a small amount of information.

Another object of the present invention isto provide a sound synthesizer which permits relatively easy extraction of thefeature parameters and operates stablyand in which differences in the spectral sensitivity among the parameters are small and the quantization accuracy of the parameters isthe same in the case of the same quantization bits.

Another object of the present invention is to provide a sound synthesizer which is excellent in interpolation characteristics for parameters used and 85; hence is able to obtain a synthesized sound of high quality with a small amount of information.

Yet another object of the present invention is to provide a sound synthesizerwhich can be produced ina relatively simple structure.

In a linear predictive analysis,the speech spectral envelope is approximated by a transfer function of an al [-pole filterwhich is given bythe following expression M:

11(z) = -- 0 0 (l) A.,,CZ7 - 1 + CCIZ + a2Z2 + a p Zp where Z = e-11A3 is a normalized angular frequency 2nfAT, AT is a sampling period, f is a sampling frequency, p is the degree of analysis, oti (i = 1, 2---. p) are predictor coefficients which are parameters for control ling the resonance characteristic of the filter and cr is the gain of the filter. Here, APM is represented by the sum of two polynominals which can be expressed as follows:

ApM = 1/2{PM + Q(Z)}... (2) P(Z) Ap(Z) - Z-ZPAp(Z-1)... (3) G(Z) Ap(Z) + Z2PAp(Z-1)... (4) (a) When the degree of analysis pis even, the expressions (3) and (4) are factorized as follows:

PM = (1 - Z) P/2 0. Z + Z 2 jI (1 - 2cos G, Q(Z) = (1 ' Z) P/2 Z2) (1 - 2 cos 0 j z + (b) When the degree of analysis pis odd, the expressions (3) and (4) are factorized as follows:

P(Z) = (1 - Z 2) (P-1)/2 + Z 2 lll (1 - 2 cosco i z QM =)/2 (1 - 2 cos 0I Z + Z 2) All (5) (6) wi and E)i in the expressions (5) and (6) are called aline spectrum pair (hereinafter referred to as LSP) and in the present invention, they are used as parameters for representing spectral envelope information.

Expressing Ap(Z) as given bythe expression (2),the' transfer function HM becomes asfollows:

2 GB 2 131659 A 2 H(Z) = - ' = a A7W 1 + (A p (Z) - i)' a (7) 1 +I(PM - 1 + QM - l} 2 The transferf unction H(Z) is also formed as a filter having two feedback loops whose transfer functions P(Z) - 1 and Q(Z) - 1, respectively. The transfer functions I'M and G0 are anti-resonance circuits and their output become 0 ato)i and wi. The frequency characteristic of Ap(Z) becomes as follows:

p/2 (COSW_ COSC0i)2 2,P/2 + sin (COSCO - Coswi)2) (8) where Z = e-jl. It appears from the above expression (8) that in a region where adjacent line spectral frequencies are closeto each other, 1Ap(Z) 12 issmall and the transfer function H(Z) exhibits a strong resonance characteristic. By changing the values of the LSP parameters coi and Oi describing the resonance characteristic of the transfer functions, an arbitrary speech spectral envelope can be obtained.

The procedure to obtain the LSP parameters is as follows; inthefirst step, auto correlation coefficients of speech wave are obtained at intervals of, for example, 10to 20 msec, in the second step, predictor coefficients (xi of the transfer function H(Z) are obtained from the auto correlation coefficients, and in the third step, the solutions of the two polynominals P(Z) and Q(Z) are obtained from the predictor coefficients on the basis of the relationship of the expression (2), thus obtaining the LSP parameters o)i and 0j. By controlling coefficients of the synthesis filterthrough utilization of the parameters representing the speech spectral envelope information, there can be obtained a filter whose transfer function H(Z) is equivalentto the speech spectral envelope. The transfer function of the feedback loop in the synthesis filter is provided in the form of a cascade connection of second-order filters, whose zeros are on a unit circle in a plane Z, as indicated by the expressions (5) and (6). Since these two second-orderfilters are identical in construction, the construction can also be simplified by multiple utilization of one second-order filter using time shared operation orwhat is called a pipeline operation. It is also possible to perform the filter operation bythe processing of an electronic computer without forming the second-orderfilters as circuits.

As described above, in the present invention the characteristics of the synthesisfilter are controlled by the aforesaid parameters o)i and Oi but, in addition to these LSP parameters coi and 6j, a fundamental frequency parameter and an amplitude parameter are employed as is the case with this kind of speech synthesizers heretofore used. Bythe fundamental frequency parameter, a voiced sound source is controlledto generate a pulse or a group of pulses of thefrequency indicated by the parameter; the output from the voiced sound source orthe outputfrom a noise source is selected depending on whetherthe soundto be reconstructed isvoiced orunvoiced; the selected outputis applied to the sound synthesis fitter; andthe magnitude of a signal onthe inputor outputside of the synthesis filter is controlled bythe amplitude parameters.The LSP parameterscoi and Gi aresubjectedto cosine transformation by parameter transforming meansto obtain -2coswi and -2cosOi, which are used as control parametersfor controlling the coefficients of the second-order filters of the sound synthesisfilter respectively corresponding to the parameters. Thecontrof parameters are interpo- lated by interpolating means in thefor of the cosine-transformed LSP parameters -2coscoand -2cosOi.Also the interpolating means may be employedforthe interpolation of the amplitude parameter. The LSP paranieterswiand 01 are excel- 7G lent in interpolatability and.the interpolation is conducted attime intervals equal to ortwice the sampling period of the original soundfor producing the parameters; for example,the LSP parameters coi and E)i are updated everyframe of 20 msec and the parameters in each frame arefurther interpolated every 125 psec. It is also possible to effectthe interpolation in the state of the LSP parameters wi and Oi and convertthem tothe control parameters.

The LSP parameters (j)i and ei are small in the amount of information perframe as compared with the control parameters for the synthesis filter for speech synthesis in the past and excellent in interpolation characteristic. Accordingly, it is suitable to transmit or store the LSP parameters wi and E)i as they are and it is also possible to convert the received or reconstructed LSP parameters coi and E)i to the control parameters forthe synthesis filter employed in other speech synthesizing systems, i.e. the PARCOR coefficients or linear predictor coefficients. In this way, the LSP parameters coi and Oi can also be used in the existing speech synthesizers. The sound synthesizer of the present invention is applicableto the synthesis of not only ordinary speech but also sounds such as a time signal tone, an alarm tone, a musical instrument sound and so forth.

Fig. 1 is a block diagram showing thefundamental construction of an embodiment of the sound synthesizer of the present invention; Fig. 2 is a block diagram showing a specific operative example of the sound synthesizer of the present invention; Fig. 3 is a circuit diagram showing an example of a f irst-order or second-orderf ilterforming a synthesis filtersection; Fig. 4A is a diagram il lustrating an example of the synthesis f ilter section in the case of the degree of analysis being even; Fig. 413 is a diagram illustrating an example of the synthesis filter section in.the case of the degree of analysis being odd; Fig. 5 is a diagram showing the relationship between the ISP parameters coi and Oi and the speech spectral envelope; Fig. 6 is a circuit diagram illustrating a specific operative example of the synthesis filter section in the case of the degree of analysis being 4; Fig. 7 is a circuit diagram illustrating a specific operative example of the synthesis filter section 3 GB 2 131 659 A 3 obtainedbyan equivalent conversion ofthecircuit shown in Fig. 6; Fig. 8 is a circuit diagram showing a specific example ofthe synthesis filter section in the case of 5 the degree of analysis being 5; Fig. 9 is a circuit diagram showing a specific operative example of the synthesis filter section obtained byan equivalent conversion of the circuit shown in Fig. 8; Fig. 10 is a block diagram illustrating an example of the synthesis filter section employingthe pipeline calculation system; Figs. 11Ato 111, inclusive, showtiming charts s[Towing the variations of signals appearing at respective parts during the operation of the filter section depicted in Fig. 10; Fig. 12 is a circuit diagram showing the casein which the filter operation achieved by the operation shown in Fig. 11 is provided by a series connection of filters; Fig. 13 is a block diagram illustrating an example of the synthesisfilter using a microcomputer; Fig. 14A is a diagram showingthe variations of powerwith the lapse of time in the casewhere a speech "ba ku o N ga-was made; Fig. 14B is a diagram showing the fluctuations in the LSP parameters wi and Oi with the lapse of time in the case wherethe speech "ba ku o N ga" was made; Fig. 15 is a diagram showing the relative frequency distributions of the LSP parameters coi and 0i to frequency; Fig. 16 is a diagram showing the relationship between the number of quantizing bits perframe and the spectral distortion by quantization; Fig. 17 is a diagram showing the relationship of the spectral distortion by interpolation to the frame length in the case of the parameters having been interpolated; and Fig. 18 is a diagram showing an example of synthesizing speech by converting the LSP parameters o)i and 01 to (x parameters.

Referring f irstto Fig. 1, the feature parameters of a speech to be synthesized are applied from an input terminal 11 to an interface section 12 every constant period of time (hereinafter referred to as the frame period), for example, every 20 msec and latched in the interface section 12. Of the parameters thus input, the LSP parameters o)i and E)i indicating spectral envelope information are provided to a parametertransform- ing section 13; and, of prameters indicating sound source information, amplitude information is applied toa parameter interpolating section and the other parameters, that is, information indicating the fundamental period (pitch) of the speech and informa- SS tion indicating whetherthe speech is a voiced or unvoiced sound are applied to a sound source signal generating section 15.

In the parameter transforming section 13, the input I-SP parameters coi and Oi are transformed into control parameters -2coswi and -2cosOi for a synthesis filter section 16, which parameters are provided to the parameter interpolating section 14. In the parameter interpolating section 14, interpolation values forthe control parameters and the sound source amplitude parameter are respectively calculated at regulartime intervals so thatthe spectral envelope may undergo a smooth change. The control parameters thus interpolated are supplied to the synthesis filter section 16, and the sound source amplitude parameter is applied to the sound source signal generating section 15. In the sound source signal generating section 15, a sound source signal depending on the features of speech is produced on the basis of the pitch information and the voiced or unvoiced sound information, and the sound sou rce signal thus obtained is applied to the synthesis filter section 16 togetherwith the interpolated sound source amplitude parameter. In the synthesisfilter section 16, a synthesized speech is producedfrom the sound source signal and the control parameters. The output from the synthesis filter section 16 is provided to a digital-analog converting section 17 and derived thereform as an analog signal at its outputterminal 18. Acontrol section 19 generates various clocksfor activating the speech synthesizer correctly and suppliesthem to the respective sections.

Fig. 2 illustrates in a little materialized form each section of Fig. 1. Everyframe period the information on thevoiced or unvoiced sound of speech is applied from the interface section 12to a voiced sound register23 and an unvoiced sound register24, and a voice frequency parameter indicating the voice pitch is stored--- in a pitch register 25. The content of the pitch register 25 is preset in a down counter 27. The down counter 27 counts down pulses of a sampling frequencyfrom a terminal 26 and every time its content becomes zero, the counter 27 presets therein the content of the pitch register 25 and, atthe same time, supplies one pulse to a gate 31. To the gate 31 are applied the outputfrom the voiced sound register 23 and an output pulse or pulses from a pulse generator 28, and when these inputs coincide, the content of a sound source amplitude register 34 is provided via the gate 31 to an adder 32. In other words, when the speech to be synthesized is voiced sound, the amplitude information is applied to the adder 32 from the sou nd sou rce amplitude register 34 every period of fundamental voice frequency of the pitch register 25, the amplitude information from the sound source amplitude register 34 being preset therein from the interpolating section 14.

In the case where the speech to be synthesized is an unvoiced sound, the output from the u nvoiced sound register 24 and a pseudo random series pulse f rom a pseudo random signal generator 36 are provided to a gate 37, and upon every coincidence of the both inputs, the amplitude information in the sound source amplitude register 34 is provided via the gate 37 to the adder 32. A sound source signal thus derived from the adder 32 is amplified, if necessary, by an amplifier 39 and then applied to the speech synthesis filtersection 16.

Inthe parameter transforming section 13,the LSP parameters coi and Oi andtheamplitude parameter are set in a register 21 from the interface section 12 everyframe period. The LSP parameters coi and Oi are applied to a parameter converter 22, wherein they are transformed to control parameters -2coswi and -2cos0j. The parameter converter 22 isformed, for example, by a conversion table of a read only 4 GB 2 131 659 A 4 memory (ROM),which is arranged so thatwhen accessed with addresses corresponding to o)i and 0j, -2coswi and -2cosOi are read out. A shift register 20 receives alternatelythe outputfrom the parameter converter 22 and the amplitude parameterstored in the register 21 and converts them to a series signal, which is applied to the parameter interpolating section 14.

In the illustrated example,the parameter interpo- lating section 14 is shown to perform a linear interpolation. Upon turning ON a switch 29, the parameters of one frame are supplied to a subtractor 30, wherein a difference is detected between the parameter and that of the previous frame from an adder 33. The difference is store in a difference value register 38 via a switch 91. Thereafter, the switch 91 is chang ed over to the output side of the difference va lu e register 38 and the content thereof is circulated. Atthis ti me, the content of the difference val ue register 38 is taken out from bit positions hig her than a predetermined bit position and supplied to the adder 33, wherein it is added to the content of an interpolation result register 92. For example, in the case of the parameter update period being 16 msec, if it is necessaryto provide interpolation parameters 128times during a frame update period,then the interpolation step width is a value obtained by dividing the difference value by 128 and this is obtained by shifting the difference value in the difference value register38 towards the lower order side by seven bits. The result of addition bythe adder 33 is provided to the interpolation result register 92 and, atthe same time, it is used as the output from the parameter interpolating section 14. In this way,there are derived from the adder 33 the values that are obtained by sequentially adding values once, twice, three time---. the shifted value of the difference register 38 to the parameter of the previous frame in the interpolation result register 92 every circulation of the difference value register 38.

In this example, the parameter interpolating section 14 is used forthe control parameter and the amplitude parameter on a time-shared basis, so that, though not shown, the control parameter and the amplitude parameter are alternately interpolated and the interpolation result register 92 is used in common to the both parameters. The amplitude parameter interpolation in the parameter interpolating section 14 is provided to the amplitude information register 34 in the sound source signal generating section 15, whereas the control parameter interpolated as mentioned above is applied to the speech synthesis filter section 16 as information for controlling itsfilter coefficient. The parameter update period,that is, the frame period, is selected to be in the range of 1 Oto 20 msec, and the interpolation period is selected to range from one to two sampling interval.The interpolation method is not limited specificallyto the linear interpolation but may be othertypes of interpolation. The point is to ensure smooth variations of the interpolated parameters.

The synthesisfilter section 16 is provided with a loop forfeeding backthe outputthrough filter circuits 41 and 42 parallelly connected each other. The filter circuits 41 and 42 are supplied with the interpolated control parameterfrom an inputterminal 44 and the outputs from the filter circuits 41 and 42 are added together by an adder 43, the outputfrom which is, in turn, added to the inputto the f ilter section 16 in an adder 45. The added outputtherefrom is fed back to the filter circuits 41 and42and,atthesametime, derived at an output terminal 55.

As each of the filter circuits 41 and42,useismade of a circuitwhich has a plurality of zeros on a unit circle in a complex plane. Thefilter circuits 41 and 42 can be both formed by a multi-stage cascade connection of f irst-order andlor second- orderf frters. In the case of forming the f ilter circuits as digital filters, use can be made of a first-order filter such, for -80 example, as shown in Fig. 3Awhich is composed of a delay circuit 51 having a delay of one sample period and an adderfor adding the delayed output and a non-delayed input, a second-order filter such as shown in Fig. 313 which is composed of two stages of delaycircuits51 and the adder 52 for adding the delayed output and the non-delayed input, and a second-order filter such as shown in Fig. 3Cinwhich the output from a multiplier 53 for multiplying the delayed output from one stage of delay circuit 51 by -2coscoi, the delayed output from two stages of delay' circuits51 and the non-delayed input are added together by the adder 52. The transfer functions of the filters shown in Figs. 3A, 313 and 3Care 1 Z, 1 -Z2 and 1-2coscoiZ+Z2 respectively. It is also possible to employ higher orderfilters.

The combination and the number of such filters depend on the degree of analysis and selected as shown in Fig. 4A or 413 depending on whetherthe deg ree of analysis is even or odd. In Fig. 4A, the degree of analysis is 10, namely, an even number and the filter circuit 41 is Constituted by a series connection of a first-order filter 56 having the transfer function 1-Z and second-orderfilters 57 to 61 each having the transfer function 1 -2cos(oiZ+Z2, and the output atthe output terminal 55 is multiplied by +112 in a multiplier 63 and applied to the series circuit. The output from the second-order filter 61 ofthelaststage and the outputfrom the multiplier 63 are added together by an adder 62 and the added output therefrom is provided to the adder43. In the filter circuit 42, the output from the multiplier 63 is supplied to a series circuit of a first- orderf ilter 64 having the transfer function 1 +Z and second-order filters 65 to 69 each having the transfer function 1-2cosOj+Z2, and the outputfrom the series circuit and the outputfrom the multiplier 63 are added together in an adder71, the added outputfrom which is applied to the adder43. The multipliers 53 of the second-o rder filters 57 to 61 are respectively given control parameters a, = -2costo, to a5 = -2cosco5 and the multipliers 53 of the second-order filters 65to 69 are respectively given control parameters bl -2cosO, to b5 = -2cos65.

Fig. 413 showsthe case wherethe degree of analysis is 11, namely, an odd number. Inthe filter circuit 41, the first- orderfilter 56 employed in the case of Fig. 4A is omitted but instead a second-order filter 72 having atransferfunction 1 _Z2 is used. In thefilter circuit42, the fi rst-order filter 64 is omitted but instead a second-orderfilter73 given a parameter b6 = 1 --- GB 2 131 659 A 5' -2cos661sused.

In the filter circuits 41 and 42the control parameters wiand E)i represent anti-resonance frequencies, at which the outputs from the filter circuits 41 and42 become 0.5. Accordingly, in the case where the anti-resonance frequencies applied to the filter circuits41 and 42 are close to each other, the output from the adder 43 becomes close too unity and the feedback loop gain approaches unity. As a consequ- ence, a high resonance characteristic appears atthe outputterminal 55. Here, col to co5 and 01 to 05 are anti-resonance frequencies which are characteristic of the speech spectral envelope information. These parameters and the spectral envelope characteristic bear such relationship as depicted in Fig. 5, from which it appears that the resonance characteristic of the spectrum can be expressed bythe spacing between adjacent parameters. These parameters have thefollowing relationship of order:

0 < 01 < 0)l < 02 <52... < 0 1 <0) i < 'C (80 The synthesizing filter has the feature that it is stable when the above condition is fulfilled.

Next, a description will be given of a specific operative example of the synthesisfilter section 16. Corresponding to theterm in the braces of the denominator in the expression (7), the following identical equations are obtainedfrom the expression (5):

p/2 PM - 1 - (l - Z) fl (1 - 2 coscaiZ + Z 2 1 i=1 p/2-1:L Z ((a, + Z.) + 17 (a,+, + Z) g (1 a Z + Z2) 80 i=l j=1 i P/2 (1 + a Z + Z?-)} jql j p/2-1:L Q(Z) - 1 = Z {(bl + Z) + (b Z) (1 + b Z + Z 2) 1+1 + j! J 85 P/2 J7 j=1 a. -2cos(g.

3. 1 b:L -2cosoi 0<wj ' Oi<7r (11) A digital filter is formed which hasanall poletransfer function approximating the speech spectral envelope given bythe expression (1) using the relationships given bythe expressions (7), (9) and (10). Fig. 6 shows the case where P = 4. In Fig. 6, parts corresponding to those in Figs. 3B to Fig. 4 are identified bythe same reference numerals. The input from the terminal 54 is added bythe adder 45to the outputfrom the adder 43, and the added output is provided to the output terminal 55 and, atthe same time, multiplied by + 112 in the multiplier63. This 112 multiplication correspondsto that in the denominator in the expression (7). The outputf rom the multiplier 63 is applied to delay means 74whose delaytime is one sampling period, i.e. the unit time. The delayed output is applied as the inputto each of the second-orderfilters 57 and 65, in which it is applied to the delay means 51, the multipliers 53 and the adders 52. In the both multipliers 53, the inputs thereto are respectively multiplied by a a, and bl, and the multiplied outputs are each applied to an adder 94 for addition with the output from the delay means 51 in each of the filters 57 and 65. The outputsfrom the both adders 94 are provided to a common adder 81 and, at the same time, applied to the adder 52 via delay means having a delaytime of one sampling period in each of the filters 57 and 65. The outputsfrom the both adders 52 are respectively applied asthe outputsfrom thefilters 57 and 65to the second-orderfilters 58 and 66 of the next stage. The filters 58 and 66 are identical in construction with thefilters 57 and 65, butthe coefficients for the multipliers 53 are a2 and b2, respectively. The outputfrom the adder 94 of each filter is applied to an adder 82 for addition with the output from the adder 81. The outputs from the adders 52 of the both filters 58 and 66 are supplied to the adder 43 for subtraction from each other, and the adder 43 is further supplied with the outputfrom the adder82.

The delay means 74 corresponds to Z outside the braces in the expressions (9) and (10), and the filters 57 and 58 each constitute a second-order filter having a transfer function 1 + Z(aj + Z), and similarlythe filters 65 and 66 each constitute a second-orderfilter having a transfer function1 + Z(bj + Z). Accordingly, the series connection of the second-orderfilters 57 and 58 realizes the third term in the braces in the expression (9), and the delay means 51, the multiplier 53 and the adder 94 in thefilter 58 realize (ai,l + Z); consequently, bythis circuit and the second-order filter 57,the second term in the braces in the expression (9) is realized, and the output is provided via the adder 82 to the adder 43.The delay means 51, the multiplier 53 and the adder 94 in the second-order filter 57 realize (a,+ Z) and the output is supplied to the adder 43 via the adders 81 and 82. In this way, the terms in the braces in the expression (9) a re realized by the second-order filters 57 and 58 and the adders 43,81 and 82. Likewise, the terms in the braces in the expression (10) are real ized by the second-order filters 65 and 66 and the adders 43,81 and 82. The expressions (9) and (10) differ in form only in that the signs of the third terms in the braces are different from each other, and on account of this difference, the sign of the input to the adder 43 differs. Accordingly, the adder 43, the second-order filters 57, 58,65 and 66, the multiplier 63 and the delay means 74 realize the expression (2), and the circuit arrangement of Fig. 6 materializes the expression (1) as a whole. In this circuit arrangement, the expressions (9) and (10) are materialized byforming the filter circuit 41 with a series connection of (P14's second-order f ilters 57 and 58 and the f i Iter circu it 42 with a series connection of (P14's second-order f ilters 65 and 66 in the feedback loop, by taking out the nodes of the second-order f ilters of the f i Iter circuit 41, that is, taps 96 and 97, from the output sides of the adders 94 to 1% obtain the total sums with the adders 81, 82 and 83. The arrangement for taking out outputs from the taps of the filter circuits will hereinafter referred to as the tap outputtype.

In Fig. 6, the second-orderfilters are arranged towards the adder 43 in an increasing order of the value j butthey may also be arranged in a decreasing 6 GB 2 131 659 A 6 order of the value j. In such a case, for example, as shown in Fig. 7, the output from the delay means 74 is provided to the second-order filters 58 and 66, the outputs from which are applied via the second-order filters 57 and 65 to the adder 43. In Fig. 7, the preceding stage of each second-o rder filter in Fig. 6 is exchanged with the succeeding stage; namely, the circuit 94for adding together the outputs from the delay means 51 ' and the multiplier 53 is exchanged with the delay means 95. The output from the delay means 74 is providedvia the taps 96 and 97 to the nodes of the second-order filters 57 and 58. In other words, the circuit arrangement of Fig. 6 is the tap output type, whereas the circuit arrangement of Fig. 7 is a tap input type. The circuit beginning with the tap 96 and ending with the adder 43 constitutes the first term in the braces of the expression (9), and the circuit from the tap 97 to the adder 43 constitutes the second term in the braces of the expression (9). The second-order filters 65 and 66 of thefilter circuit41 are also similarlyformed. In connection with the filter circuit 41,the outputfrom the delay means 74 is multiplied by -1 in a multiplier98 to materializethe minus signforthe thirdterm in the braces of the expression (9).

In the case where p is odd, thefollowing identical equation is obtained from the expression (8) corresponding to theterm in the braces of the denominator in the expression (7).

(P-3)/2:i Z2 p(Z) Z ga. + Z) + + Z) J, + a.Z + - Z (pT1)/2 (1 + a j Z + Z2 j =91 (P-1)/2 QM - 1 = Z{(bl + Z) (b Z) :L+1 + :L X 17 (1 + b.Z + Z2)) j=1 j a. -2cos b ?CO'S 0.<' 0 1 (12) (14) As in the case of p bei rig even, two types of digital filters respectively called the tap outputtype and the tap inputtype are materialized in such forms as shown in Figs. 8 and 9 from the relations of the expressions (7),(12) and (13). In Figs. 8 and 9, it is assumedthat p is 5. In Figs. 8 and 9, the first-order filter72 correspondsto Z in thethird term in the braces of the expression (13) and the second-order filter73 is to obtain such a characteristic that the products of the transfer functions (1 + blZ + Z') and (1 + b2Z + Z) of thefilters 65 and 66 is multiplied by (b3 + Z).

As Will be understood from Figs. 6to 9, the +112 multiplier 63 and the delay means 74 may also be disposed at any places in thefeedback loop. Since the second-orderfilters are of the same type, it is possible to simplify hardware byforming the circuit arrangement so thatthe so-called pipeline operation is effected by using, on a time-division multiplex basis, one multiplier 53,the plurality of adders 52 and 94 and the plurality of delaymeans51 and95rviaking up one seco nd-order filter. Fig. 10 illustratesthe case where the example of the filter shown in Fig. 21 is arranged to conduct the pipeline operation. In this example,p= 10, and an operation of a setof parameters applied from the interpolating section is completed with a period of 176 clocks. In Fig. 10, parts corresponding to those in Fig. 12 are marked with the same reference numerals. The input side of a 16-bit static shift register 74, which performs thefunction of the delay means 74 is changed over by a switch S, between the output side of the shift register itself and the output side ofthe adder 45. Am u Itipi icand input side ofthe multiplier 53 and the input side of the adder52 are changed over by a switch S2tO the output side of the shift register 74, the output side of a (27-d)th shift stage counted from the input of the shift register 74 and the output side of a 31-bit shift register 101, d being an operation delay of the multiplier 53. The multiplier 53 is connected atone end to the output terminal 55 and the Input side of the adder 94 and derives at the other output end the multiplicand input delayed by 22 clocks, which is provided to the (154 + d)-bit shift register 51. The outputfrom an adder81 is fed backto the input side thereof via agate 102 and a 16-bit shift register 103, performing a cumulative addition through the adders 81 and82in Fig. 12. The gate 102 is opened only in the time interval between d+2 and 145+d. One input side of the adder 43 is changed over by a switch S3 between the output sides of the adders 52 and 81, and the other input side of the adder 43 is changed over by a switch S4 between the output sides of a 16th and a (d+ 1)th shift stages of the shift register 101. The input side of the shift register 101 is changed over bya switch S5 between the output sides of the adders 43 and 52.

The switches S, to S5 are each connected to the fixed contact side, during one operation period, that is, 176 clocks, for a clock period indicated by numerals labelled atthe fixed contact. The shift registers 51, 95, 101 and 103 are respectively (1 54 bit, (1 75-d)-bit, 31-bit and 16-bit dynamictype ones and are always supplied with shift clocks. The broken line inputto each of the adders 43,45,52,81 and 94 indicates the timing of the operation boundary of each parameter; for example, % indicates a repetition every 16 clocks and an operation delay of each adder is selected to be one clock. Fig. 11 is a timing chart of the operation of each pa rt in Fig. 10, Fig. 11 a showing the timing of the clock, Fig. 11 B the inputs of the coefficients ai, bi and Ato the multiplier 53 from the inputterminal 44, Fig. 11 C the multiplicand of the multiplier 53, Fig. 11 D one in put to the adder 94from the mu Itiplier 53, Fig. 11 E the other input to the adder 94, Fig. 11 F the outputfrom the adder 94, Fig. 11 G the outputfrom the adder 81, and consequentlythe content of the register 103, Fig. 11 H the inputto the adder 52 from the shift register 95, and Fig. 111 the output from the adder 52. Fig. 12 shows these inputs and outputs in the form of signals appearing atthe respective parts in the case wherethe second-order filters are cascade- connected.

As shown in Fig. 11, in the period between clocks 0 and 16, a coefficient al(t) and a multiplicand xl(t) are multiplied in the multiplier 53 to effeetthe multiplica- tion in the second-order filter 57 in Fig. 12, and the 7 GB 2 131659 A 7 result of multiplication is obtained from a dth clock. In the period between clocks 16 and 32, as shown in Figs. 11 B and 11 C, a coefficient bl (t) and a multiplicand y, (t) are multiplied to perform the multiplication in the second-orderfilter 65. The multiplicand xl (t) is delayed bythe shift register 51 along with 22 bits of the multiplier 53 by (176+d) clocks, so that as shown in Fig. 11 E, a multiplicand xl(M) is appliedto the adder94fromthe dth clock and added with the output alx, derived from the multiplier 53 atthattime, and the added outpLtxl'(t) is provided via the adder 81 to the shift register 103 for accumulation. That is, the outputfrom the adder 81 is supplied to the signal system of the adders 81,82,... in Fig. 12.

The output from the adder 94 is also provided to the (175-d)-bit shift register 95, as shown in Fig. 11 H. Accordingly, in the period between the clocks 0 and 16, the outputfrom the shift register is xl'(t-1), as shown in Fig. 11 H, andthis output is added with the multiplicand xl(t) in the adder 52,the OUtPUtX2W from which is applied asthe inputto the second-order filter 58 in Fig. 12. The OUtPUtX2(t) fromthe adder 52 is provided via the shift register 101 to the multiplier 53. As shown in Fig. 1 1C, the Output X2W is multiplied by the coefficient a2(t) in the multiplier 53 in the period between clocks 32 to 48. Priorto this multiplication, bi (t) andy, (t) are multiplied, as described previously, and the multiplied output is similarly processed, therebyto obtain the outputy2(t) from the second- orderfilter65 in the period between clocks 48 and 64. In this way, the multiplication of the coefficient a and the multiplicand x and the multiplication of the coefficient b and the multiplicand y are carried out alternately every 16 clocks, and the multiplied results are applied to the shift register 51, as indicated by alxl, blyl, a2X2, b2Y2,... in Fig. 11 D. Further, the second-order filters 57,58,59,60 and 61 respectively derive therefrom Xl'(t), X2t), X3'(t), X4'(t), x5(t) and X2(t), x3(t), x4(t), x5(t), x6(t), which are provided to the shift registers 95 and 101. Similarly, yl'(t) to y5'(t) and Y2(t) to y6(t) are respectively obtained from the second-orderfilters 65 to 69, and these outputs are applied to the shift registers 95 and 101 alternately with xt) and x(t), respectively. In the period between clocks 145 and 161, the outputy6 derived fromthe adder 52 atthattime and X6 in the shift register provided previously are subtracted onefrom the other in the adder43, and (X6-Y6) is supplied via the switch S5 to the shift register 101, wherein it is delayed by (d+l) clocks. The delayed output is taken outfrom the switch S4for input to the adder 43 in the period between clocks 147+d and 163+d. The output yielded from the shift register 103 atthattime is provided to the adder 43 via the adder 81 and the switch S3. The output from the adder 43 at that time becomes the output from the adder 43 in Fig. 12 and this output is applied to the adder 45, wherein it is added with the input at the terminal 54 to provide Z(t). The added output Z(t) is supplied to the register 74, wherein it is delayed by the delay means 74 in Fig. 12. The delayed output is applied to the multiplier 53 and atthattime the coefficientA is provided as an amplitude interpolation output atthe terminal 44 and A. Z(t) is derived from the multiplier 53 atthe output terminal 55. This multiplication is performed in the case wherethe outputfrom thesynthesis filter section 16 is multiplied bythe amplitude information A in a multiplier 104 in Fig. 12. From the shift register 74 is taken out an output Zffi/2 having shifted down by one bit and this is taken out via the switch S2tO the multiplier 53 as Z(M)12, that is, x(t) and y(t), in the next subsequent operation period fora new set of the parameters. The output at the output term ina 155 can also be obtained as parallel outputs through an output buffer 105 of a static shift register.

The pipeline operation described above is also applicableto othertypes of synthesisfilter section 16. Furthermore, as will be appreciated from the arrangement of Fig. 10, the filter operation can be achieved by the addition, multiplication and delay, so that this filter processing can also be effected using a microcomputer. For example, in Fig. 13, by successively reading out, interpreting and executing programs in a program memory 107, a central processor unit 106 loads therein from an input port 111 a sound source signal and control parameters respectively applied from the sound source signal generating section 15 and the interpolating section 14to terminals 108 and 109, and- the central processor 106 sequentially performs the operations described previously with regard to Fig. 11. A read-write memory 112 is used instead of the registers 51,74,95, 101, 103 and 105 in Fig. 10. The results of the operations are written in the read- write memory 112 and read outtherefrom at suitable timing to perform operations. The output thus obtained is applied from an output port 113 to the outputterminal 55. The central processor 106,the memories 107 and 112 and the ports 111 and 113 are connected to a bus 114.

By anyone of the abovesaid methods the output from the synthesis filter section 16 is obtained. The output is converted by the D-A converting section 17 in Fig. 2to an analog signal to provide a speech output. In the D-A converting section 17, if the input thereto is a serial signal, then it is applied to a shift register 115 and the content of the shift register 115 is converted by a D-A converter 1 16to analog form.

As described previously, the LSP parameters wi and Bi in the speech feature parameters used in the present invention can be obtained by obtaining the solutions of the expressions (5) and (6). In Figs. 14A & 14B there are shown the results of analysis of a speech "bakuoNga" using the LSP parameters col and %. In Figs. 14A and 1413, the abscissa represents time t, in Fig. 14Athe ordinate represents power, and in Fig. 14B the ordinate represents normalized angular frequency. Seeing instantaneous points in Fig. 1413, thefrequency rises in the order of parameters E),, to,, E)2, 0)2,... 05, C05Ahis orderdoes not change and the parameters Oi and o)i do notcoincidewith each other in oneframe. Accordingly, it is guaranteed thatthe synthesizing filter section 16 is always stable. The frequency distributions of the LSP parameterseiand col are shown in Fig. 15, in which the abscissa represents normalized angularfrequency f and the ordinate the relative frequency D. As shown in Fig. 15, each parameter is not distributed over a wide frequency band but restricted to a relatively narrow frequency band, so that the LSP parameters col and Oi can be quantized in connection with thefrequency 8 GB 2 131659 A 8 range in which they are distributed.

The LSP parameters coi and E)i are little in quantizing distortion. Fig. 16 shows a spectral distortion Ds of a synthesized speech when various parameters were quantized variously, the abscissa representing the number of quantizing bits B perframe and the ordinate the spectral distortion Ds. The line 117 shows the case where in consideration of onlythe parameter distribution, the PARCOR coefficient is quantized linearly only in the coefficient was distributed; the line 118 shows the case where the number of quantizing bitsforthe PARCOR coefficientwas increased in consideration of the spectral sensitivity in addition to the parameter distribution in the case of the line 117, especially in the case of markedly ' affecting the spectrum; the line 119 shows the case where the LSP parameters wi and Oi were quantized in consideration of only the parameter distribution; and the line 121 shows the case where the LSP para- meters wi and Oi were quantized in consideration of the parameter distribution and the spectral sensitivity. Itwill be seen from Fig. 16 that in the case of using the same number of quantizing bits, the spectral distortion becomes smaller in the orderof the lines 117,118,119 and 121. Since the lines 119 and 121 are close to each other, the LSP parameters o)i and Oi are not so much affected in spectral distortion even if the spectral sensitivity is nottaken into account. Accordingly, since it is sufficientto perform the quantiza- tion taking into consideration the parameter distribution range alone, the quantization is easy. The value that the nu mber of quantizing bits perframe at which the spectral distortion is 1 dB in the case of the line 119 is divided bythat number of quantizing bits in the case of the line 117 is 0.7. Similarly, the ratio of the number of quantizing bits perframe atwhich the spectral distortion is 1 dB between the lines 118 and 121 is 0.8. From this, itwill be understood thatthe LSP parameters coi and Oi are excellent. One dB is a difference limen of the spectral distortion of a synthesized speech.

Fig. 17 shows interpolation characteristics, the abscissa representing a frame length Tf and the ordinatethe spectral distortion Ds. Fig. 17 shows the spectral distortion of a synthesized speech in the case where a frame in which an original speech was analyzed in 10 msecwas used asthe reference, the frame length was increased to 20 to 70 msec and parameterswere interpolated every 10 msec. The line 122 shows the case where use was made of the PARCOR coefficients, and the line 123 shows the case where use was made of the LSP parameters coi and 6j. Aswill be seenfrom Fig. 17, inthecase ofthe same distortion, the frame length Tf can be made longer by the LSP parameters than the frame length Tf by the PARCOR coefficients, that is, the parameter u pdate period can be increased, so that the entire amount of information can be reduced bythat. In addition, since the LSP parameters are smallerthan the PARCOR coefficients in the number of bits perframe, as seen from Fig. 16,the amount of information forthe same distortion may be small by the product of the reduction ratios in Figs. 16 and 17; namely, in the case of the LSP parameters, the amount of information may be about 60% of that in the case of the PARCOR coefficients.

In the case of employing the LSP parameters, it is meaningless as in the cases of other narameters that they are interpolated with a shorter period than the sample period of the original speech used in the making of the parameters. Experiments revealedthat the interpolation period might be abouttwice or less the sample period of the original speech, butthat when theformerwas aboutfourtimes the latter, noiseswere introduced to make the synthesized speech indistinct. Accordingly, it is preferredthatthe interpolation period be equal to ortwice the original speech sampling period.

As has been described in the foregoing, the LSP parameters are relatively easyto automatically extract, and consequently can be extracted on a real time basis. Furthermore, the LSP parameters are excellent in the interpolation characteristic and small in deviation of the qunatizing. characteristic and permits transmission and storage of speech in a small amount of information. In the speech synthesis, speech of high quality can be reconstructed and synthesized with a small amount of information, and as long asthe relationship of the expression (8) holds true, the stability of the synthesizing filter is guaranteed.

In Fig. 2, it is also possibletowiden the spectrum by generating from the pulse generating section 28 a train of pulse groups, such asthe Barkerseries, instead of the pulse train. The interpolating section 14 may also be provided atthe preceding stage of the parameter transforming section 13. Namely,the LSP parametersfrom the interface section 12 may also be subjected to the cosine transformation in the para- meter transforming section 13 after being interpolated. In this case,the use of a read only memory is uneconomical since its memory capacity need be enormous; accordingly, it is preferred to perform parameter conversion using an approximation op- eration of the cosine ratherthan using the read only memory as described in the example of Fig. 2. In Fig..2, the information indicating whetherspeech is a voiced or unvoiced sound is entered and loaded in the voice sound register 23 and the unvoiced sound register 24, butthis information need not always be provided. That is, a detector circuit is provided for detecting whetherthefundamental period parameter applied to the pitch register 25 is zero or not; in the case of detecting zero, the sound is decided to be an unvoiced sound and the gate 37 is opened; and in the case of othervaluesthan zero, the sound is decided to be a voiced sound and the gate 31 is opened. The control by the amplitude parameter may also be effected in connection with the outputfrom the filter section 16, as described previously with respect to the embodiment of Fig. 12.

In the foregoing, as the synthesis filter, use is made of a filter which includes in the feedback circuit the means for connecting in series a plurality of f irst- order and second-order filters of different coefficients, each having the zero on a unit circle, through utilization of the LSP parameters. Howeverthe synthesisfilter need not always be limited specifically to such a filter and the speech synthesis may also be effected by transforming the LSP parameters to some 1 9 GB 2 131 659 A 9 othertypes of parameters and using otherfilters. For example, as shown in Fig. 18 in which parts corres ponding to those in Fig. 1 are identified bythe same reference numerals, the fundamental period para meter in the feature parameters applied to the interface section 12 is provided to the sound source signal generating section 15, and the amplitude parameter is supplied to the interpolating section 14.

The amplitude parameterthus interpolated is applied 1 o to the sound source signal generating section 15, in which it is processed as described previously in respect of Fig. 2, providing a sound source signal to the synthesis filter section 16. The LSP parameters are supplied to an LSP parameter transforming section 124, in which they are transformed to other types of parameters, such as an cc parameter, PARCOR parameter orthe like. For example, from the LSP parameters are obtained polynominals P(Z) and Q(Z) usingthe expression (5) or (6), and from the polynominalsthe predictor coefficients (xi of the transfer function H(Z) are obtained using the express ions (1) and (2). By interpolating thethus obtained predictor coefficients (xi in the interpolating section 14 as required, the characteristics of the sound synthesis fitter section 16 are controlled. Thefilter section 16 isformedfor example, asa cyclicfilter, in which, as shown in Fig. 18, the sound source signal from the soundsource signal generating section 15 is made cr-fold by a multiplier 125 and applied to an adder 126 for subtraction from the output of an adder 127 and the outputfrom the adder 126 is provided to the outputterminal 55. The outputthus derived atthe outputterminal 55 is applied to a series circuit of delaycircuits D, to Dp, each having a delaytime of one sample period. The outputs from the delay circuits D, to DP are respectively multiplied by coefficients (xl to (xpfrom the interpolating section 14 in multipliers M, to MP, The multiplied outputs are sequentially added and then added together in the adder 127.

Itwill be apparentthat many modifications and variations may be effected without departing from the scope of the novel concepts of this invention.

Claims

1. A sound synthesizer comprising:

a sound source signal source for generating a sound source signal; A LSP parameter source for generating LSP para meters; parameter tra nsfo rmi ng means fortransforming the LSP parameters to control parameters of a type different from the LSP parameters; and a sound synthesis filter section supplied with the sound source signal and controlled bythetrans formed control parameters in its characteristics.

2. A sound synthesizer according to Claim 1, wherein the sound source signal source is composed of a fundamental period sound source controlled by a fundamental period parameterto generate a pulse or a pulse group of the period indicated bythe parameter, a noise source for-generating random pulses, and select means for selectively taking outthe outputfrom thefundamental period sound source or the outputfrom the noise source depending on whether a sound to be synthesized is a voice or 130 unvoiced sound.

3. A sound synthezier according to Claim 1 or 2, further comprising amplitude control meansfor controllingthe magnitude of a signal atthe inputor outputside of the sound synthesis filter section byan amplitude parameter.

4. A sound synthesizer according to Claims 1 or 2 wherein said pa ra meter transforming means comprises means fortransforming said LSP parameters wi, Oito predictor coefficients ot, and wherein said sound synthesis fi Ittr means comprises first adder means one input of which is supplied with said sound source signal, a cascade connection of a plurality of unit time delay means, the input of said cascade connection being connected to the output of said first 11 adder means, a plurality of multiplier means each for multiplying the output of one of said unit time delay means and the corresponding one of said predictor coefficients ocl, and second adder means for sum- ming up all the outputs from said multiplier means and supplying the sum to another input of saidfirst adder means.

Printed for Her Majesty's Stationery Office byTheTweeddale Press Ltd., Berwick-upon-Tweed, 1984. Published atthe Patent Office, 25 Southampton Buildings, London WC2A lAY, from which copies may beobtained.

GB 2 131 659 A 11

4. A sound synthesizer according to anyone of Claims 1 to 3, wherein the parameter transforming means is means fortransforming the LSP parameters to predictor coefficients, and wherein the sound synthesis filter section is a cylcic. digital filter.

5. A sound synthesizer in which a sound source signal and control parameters for controlling the characteristics of a filter are applied to a synthesis fitter section andfliter coefficients of the synthesis filter section are controlled by the control parameters to obtain a sythesized sound signal, characterised in thatthe synthesisfilter section is composed of second- order filter means serving as second-order filters, each having the zero on a unit circle in a complex plane, meansfor cascade-operating such second-order filter means of different coefficients, andfeedback means forfeeding backthe outputfrom the synthesisfilter section to the input side thereof through two kinds of such cascade-operating means.

6. A sound synthesizer according to Clairn 5, wherein the sound source signal source is composed of afundamental period sound source controlled by a fundamental period parameter to generate a pulse or a pulse group of the period indicated by the parameter, a noise course for generating random pulses, and select means for selecting taking out the outputfrom the fundamental period sound source or the output from the noise source depending on whether speech to be sythesized is a voiced or unvoicedsound.

7. A sound synthesizer according to Claim 5 or 6, further comprising amplitude control means for controlling the magnitude of a signal at the input or output side of the synthesis filter section by an amplitude parameter.

8. A sound synthesizer according to anyone of Claims 5 to 7, wherein the second-orderfilter means is composed of first delay means for delaying the inputfor a unittime, first adder means supplied with the delayed output and the output from the synthesis filter section, second delay means for delaying the output from the first adder means fora unit time, multiplier means for multiplying the output from the first adder means and the coefficient, the second adder means for adding togetherthe multiplied output, the outputfrom the second delay means and the input to the second-order filter means to provide the output from the second-orderf ilter means.

9. A sound synthesizer according to anyone of Claims 5 to 7, wherein the second-order filter means is composed of first delay means for delaying the input to the second-order filter means fora unit time, multiplier means for multiplying the input to the second-orderfilter means, second delay means for delaying the added output fora unit time, and second adder means for adding the output from the second delay means and the inputto the second-order filter means to provide the output from the second-order filter.

GB 2 131659 A 10 10. A sound synthesizer according to anyone of Claims 5 to 9, wherein the second-orderfilter means isformed as a second-orderfilter circuit; a plurality of such second-orderfilter circuits of different coeffi- cients are cascade-connected to form the cascadeoperating means; and a pair of cascade-connected second-orderf ilter circuits of different filter coefficients form the two feedback means.

11. A sound synthesizer according to Claim 8 or 9, wherein the second-order digital filter means is formed as a second-order digital filter circuit; and the second-order digital filter circuit is used on a multiplex basis by a pipeline operation system by operating the filter circuit a plurality of times within a unit time and changing the coefficient of the filter circuit for each operation.

12. A sound synthesizer according to anyone of Claims 5 to 9, wherein the filter means and the cascade-operating means are formed by operating means for performing filter processing by interpreting and executing a program.

13. A sound synthesizer according to anyone of Claims5to 12, further comprising parameters transforming means for obtaining the control parameters by the cosine transformation of parameters for controlling the charateristics of the synthesis filter section.

14. A sound synthesizer according to anyone of Claims 5 to 13, further comprising interpolating means for interpolating the control parameters and supplying them to the synthesis filter section.

15. A sound synthesizer according to Claim 13, which further comprises interpolating means for interpolating the parameters representing the char- acteristics of the synthesis filter section, and wherein the interpolated parameters are subjected to cosine transformation bythe parameter transforming means.

16. A sound synthesizer according to Claim 14 or 15, wherein the interpolation period in the interpolating means is equal to or twice the sample period of an original sound signal.

17. A sound synthesizer according to Claim 14,15 or 16, wherein the interpolating means is used on a multiplex basis for the interpolation of an amplitude parameter.

18. Sound synthesizing method, wherein sound source signal parameters representing a sound source signal and control parameters for controlling characteristics of filter means are supplied to a medium, the sound source signal is generated in accordance with the sound source signal parameters from the medium, the sound source signal isfed to thefilter means while the characteristics of the filter means being controlled bythe control parameters from the medium thereby producing therefrom a synthesized sound signal, characterized in that; a spectral envelope of original sound is approximated by a transferfunction H(Z) of said filter means expressed by :m H(Z) = p OZ) = 01 + a 1 Z + a 2 Z 2 + + a ZP where Z = e-j', a being a constant, a normalized angularfrequency 2TifAT, AT a sampling frequency, f frequency, p a degree of analysis, and ai (i = 1, 2,... p) predictor coefficients, said Ap(Z) isfurther expressed as a sum of two polynominals P(Z) and Q(Z) by A 2 P(Z) l(P(Z) + Q(Z)) P(Z) AP(Z) - Z.ZPA p (Z-1) Q(Z) AP(Z) + Z.ZPA p (Z-1) said polynominals are each factorized and angular frequencies which renderthe polynominals zero are used as said control parameters. New or amendments to claims filed on 3011184. Superseded claims 1-18. New or amended claims:- CLAIMS 1. A sound synthesizer comprising a sound source signal source for generating a sound source signal, an LSP parameter source for generating LSP parameters {col} and {01} expressed in angular frequencies respectively allowing the roots of polynominals P(Z) and Q(Z) assuming zero and defined by the following expressions:

KZ) - Ap (Z) - ZP+1AP (Z-1) Q(Z) - Ap (Z) + U+1AP (Z-1) AP (Z), - 1 + - 1Z + -- 2z2 +... + ot PZP where {oc j} are predictor coefficients determined from each predetermined numberof samples of soundwave signals, parameter transforming means fortransforming the LSP parametersto control parameters of atype different from the LSP parameters, and a sound synthesisfilter means having a transferfunction H(Z) = cr/Ap(Z), said sound synthesis filter means being supplied with the sound source signal from said sound source signal source while the characteristics thereof are controlled bythetransformed control parameters.

2. A sound synthesizer according to Claim 1, wherein the sound source signal source is composed of a fundamental period sound source controlled by a fundamental period parameterto generate a pulse or a pulse group of the period indicated bythe parameter, a noise sourcefor generating random pulses, and select means for selectively taking outthe outputfrom thefundamental period sound source or the outputfrom the noisesource depending on whethera sound to be synthesized is a voice or unvoiced sound.

3. A sound synthesizer according to Claims 1 or 2, further comprising amplitude control means for controlling the magnitude of a signal atthe input or output side of the sound synthesis filter means by an amplitude parameter.