JPS5840596A - Sound source circuit - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a rounding sound source circuit that provides f source data corresponding to a sound source waveform to a digital filter (grid filter).

In general, in speech analysis and synthesis methods such as linear predictive coding, the resolution of the sound source is 8 kHz (125μ5ee
)6 or 10kHz (100μI@-). Although this poses no problem in terms of sound quality for voices, the resolution is coarse for scales such as melodies and songs, which poses a problem in terms of sound quality.

FIG. 1 shows an example of a conventional sound source waveform. Here, [oI mark is the sampling point. As can be seen from the figure, the pitch value has the resolution of the sampling frequency. If the sampling frequency is 8kHz, the value of the sound source is 125μ
The unit is m.

Taking each scale shown in 91 below as an example, the maximum error is 4.7%, and the permissible error for human auditory discrimination for scales is 1- or less, so speech synthesis for melodies and songs is difficult. There is a problem with the sound quality.

Table 1 The ninth way to eliminate the sound quality problems mentioned above may be to increase the pitch resolution of the sound source, but if the digital filter (sound source data calculation circuit) installed after the sound source circuit is 8 kHz or Due to restrictions such as those specified for 10 kHz, the above resolution cannot be increased without modifying this digital filter.

The present Ij1 was developed in view of the above-mentioned circumstances, and by using two or more similar waveforms that differ only in phase as sound sources in a speech synthesis circuit based on the analysis and synthesis method, high-quality melody can be achieved using a conventional digital filter. This is an attempt to provide a sound source circuit that dances by obtaining 7' twos, songs, etc.

An embodiment of the present invention will be described below with reference to the drawings. Second
The figure shows an example of the sound source waveform of the same embodiment, in which the sound source is a triangular wave. FIG. 2(a) shows the basic waveform when the sampling frequency is 8 kHz x 5. Same figure (
In b) to (f), the sampling frequency is 8 kHz and the phase of the sound source waveform is shifted by 115 of the sampling period of 8 kHz with respect to the basic waveform of FIG.

#! Figure 3 shows the configuration of the same embodiment in a professional manner.
b, so 11. ~111 is a ROM (R@ad 0nly M@mory) that stores sound source waveform data. ROMJJ stores the sound source waveform data (actually, the digital data of each sampling point indicated by l'-J marks) in FIG. 2 (,),
Similarly, the r-ta of the sound source waveform shown in FIG. 2(f) (actually digital data at each sampling point indicated by a "mouth") is stored in the ROM JJf. 12 is RO
M J 1, ~11. This is an address code for specifying addresses. 13 is a counter that supplies address data to this address decoder 12, and this counter p13
It is reset by a match output from the u comparator 14. 1
5 is a data register that stores nine coarse pitch data (f number) obtained by analysis. Therefore, when pitch data n (integer) is stored in this register, the counter 3 counts from 10'' to @n1, and the address decoder 12 uses the count 1 as address data.
supply to.

16 is mold pitch data obtained from analysis (0 to 415
), IF is register 6 with
ROM 11 according to the stored data.

The output of this multiplexer 17 is a multiplexer that selects the output of 11f. Therefore, the input to this filter is the RO selected by the data in data register 6.
M output.

The fourth factor shows the pitch period when using the sound sources shown in Figure 2 (b) to (f), where the pitch value is n and the sound source waveforms used are (b) to (f). , pitch 4
3 2 1 value un + m s + Otori-s T n-s *
(n-s) 1 It will have an accuracy 5 times higher than the conventional precision fO 0 In terms of the circuit, conventionally only the pitch value was used as sound source data, but in the present invention, the pitch value and the sound source used are Waveform data is required, but among the sound sources (b) to (f) located between n and @n±1#, it is necessary to obtain the sound source frequency # to obtain the actually necessary scale frequency. Since it is possible to select sound source waveform data that is close to the sound source waveform, the error is at most 1.2- for the musical scale as shown in Figure 2 below, and the sound quality is greatly improved.

? 2 Note that the present invention is not limited to the above-mentioned embodiments, and can be applied in various ways.For example, the present invention is not limited to the above-mentioned embodiments, and can be applied in various ways. By adding or subtracting an arbitrary constant to the waveform using an adder/subtractor, it is possible to create sound source waveforms with different phases.

As explained above, according to the present invention, a plurality of ivy resource waveforms with different phases are prepared and these are selected, so that the error for each scale can be suppressed within the range of permissible error for human auditory discrimination. However, it is possible to use conventional digital filters (
It is possible to provide a sound source circuit which has advantages such as being able to obtain high-quality melodies using a lattice type filter.

[Brief explanation of the drawing]

Figure 1 shows an example of a conventional sound source waveform, Figure 2 (,)
-(1) are diagrams showing examples of sound source waveforms according to an embodiment of the present invention, FIG. 3 is a block diagram showing the same configuration, and FIG. 4 is a sound source waveform diagram obtained by the same configuration. 11-111...ROM, JJ...Address Deco Ichijime, 13...Kakunta, 14...Kono Arm Lator,
16.16...Data Renosta, 11...Multiple Brefusa. Applicant's agent Patent attorney Hikotsubo Suzue

Claims (3)

[Claims] (1) A sound source circuit characterized by comprising data generation means for sound source waveforms of the same type having different phases, and data selection means for each of the sound source waveforms. (2) The sound source circuit according to claim 1, wherein the phase shift between the sound source waveforms is equal to or less than the sampling period of the sound source waveforms. (3) The sound #A circuit according to claim 1, wherein the output of the sound source waveform data selection means is input to a digital filter for obtaining a sound waveform from the sound source waveform.