JPS58196594A - Musical tone synthesizer - 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 musical tone synthesis device using a phoneme segment editing synthesis method, and in particular, by storing minimum pitch data, it is possible to create high-quality synthesized musical tones that are closer to natural instrument sounds. The object of the present invention is to provide a musical tone synthesis device that can be obtained.

In general, various voice synthesis methods have been attempted as musical tone synthesis devices, and in particular, when considering the application of a phoneme segment editing synthesis method that obtains a desired musical tone signal by connecting phoneme segments one after another, it is difficult to use a single musical instrument. A method is adopted in which data obtained by converting a musical tone of a certain pitch into PCM is stored in memory, and when the data is used as a musical tone of a different pitch, the sampling period of the converting means is changed and output. Since the envelope of the rising and falling parts of musical tones cannot be reproduced, the range of pitch variation is small, and a large amount of data memory is required to store the pitch data. In other words, with the above-mentioned method of outputting by changing the /h sampling period, if you try to share a single pitch data for each pitch within one octave, it will not be possible to reproduce the envelope of the rising and falling parts of the musical tone. A synthesized musical sound with a pitch far from the pitch stored as pitch data will be far from the sound of a natural instrument. Therefore, in order to obtain a synthesized musical sound with a quality close to that of a natural instrument, it is necessary to combine multiple pitches within one octave. It is necessary to store the data, which has the drawback of requiring a large amount of data memory to store the pitch data.

The present invention solves these conventional drawbacks,
Information regarding the number of repetitions that approximates at least the vertical envelope of a natural musical tone when synthesizing musical tones of different pitches of the same instrument using phoneme data required to synthesize musical tones of a specific pitch of a certain musical instrument It is structured so that it also refers to .

The present invention will be described below with reference to drawings of embodiments. FIG. 1 shows an embodiment of the present invention. In FIG. 1, after information regarding designation of the type of musical instrument and pitch is input to the input section 1, this information is output to the control section 2. Ru.

Based on the above-mentioned information, the control unit 2 reads out information regarding the address, data length, repetition, and Dm sampling period regarding the phoneme piece data to be output from the table memory 3, and outputs the phoneme from the phoneme piece memory 4 according to these information. After reading out one piece of data and outputting it as a digital musical tone signal to the DMU converter 5 at a predetermined DMU sampling period and outputting it a predetermined data length for a predetermined number of repetitions, the information in the table memory 3 that has already been read out is output. Perform the operation to read the next information, and continue the above-mentioned operation until a signal for termination is sent from the input section. In the above,
The D music conversion unit 6 converts the digital music η output from the control unit 2.
Converts the signal to an analog musical tone signal and outputs it.

Next, regarding the repetition information and the D sampling period information in the table memory 3, the following will be described in Section □2.
I'll explain it with a quick explanation. Figure 2 a Number of repeats n1+
n2s n3... are each 1t 2,3t
...and the D sampling period 2 is Zo/
It becomes z.

However, n4 can be considered to be infinite repeatedly until the end signal is received. Also, here, it is assumed that FIG. 2a stores the envelope of the original sound. In this way, even when the frequency ratio is not an integral multiple, n,, n2゜n5...
The one that most closely approximates the envelope of the original sound is selected by combining the number of repetitions of phoneme pieces at the rising edge, such as:

The same thing as described above can be done for the falling portion as well. In other words, FIG. 2 means that musical tones of the same musical instrument with different pitches can be synthesized using data that differs only in the number of repetitions and the D sampling period of the information in the table memory 3. Therefore, since the envelope number of the rising and falling parts can be expressed by the number of repetitions, it is possible to obtain a synthesized musical sound with a quality close to that of a natural musical instrument, and even for pitches that do not exist as natural musical sounds, it is possible to synthesize sounds that are typical of that instrument. can do.

Note that for wind instruments and the like, the envelope at the falling edge has only a small effect on the quality of the synthesized sound, so there is no need to particularly control it.

As described in detail, according to the present invention, a phoneme memory that stores phoneme pieces, a control section that reads phoneme pieces from the phoneme piece memory and outputs them as digital musical tone signals;
an input unit for instructing the control unit to specify the type and pitch of a musical tone to be output; and an input unit for the control unit to read a phoneme from the phoneme memory, and input the data length, number of repetitions, and output itJ of the phoneme to be output; ]; and a Dm conversion unit that converts the digital musical tone signal output from the control unit into an analog musical tone signal,
The table memory stores digital musical tone signals to be output from the control unit when synthesizing musical tones of different pitches of the same instrument using phoneme data required for synthesizing musical tones of a specific pitch of a certain musical instrument. The phoneme memory is configured to refer not only to information regarding the time period of the natural musical tone, but also to information regarding the number of repetitions that approximate at least the rising envelope of a natural musical tone. This has the advantage that the necessary phoneme data requires only minimal information on pitches, and the compression ratio of the data necessary for musical tone synthesis can be increased.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the musical tone synthesis apparatus of the present invention, and FIG. 2 is an information waveform diagram of the main parts of the apparatus. 1...Input section, 2...Control section, 3...
...Table memo 1c4... Phoneme piece memory,
5...DM conversion section. Name of agent: Patent attorney Toshio Nakao and 1 other person No. 1
figure

Claims (1)

[Claims] A phoneme memory that stores phoneme pieces necessary for synthesizing a musical tone of a specific pitch of a certain musical instrument, and a phoneme piece memory that reads out the phoneme pieces from the phoneme piece memory and outputs them as digital music and signals. a control section; an input section for instructing the control section of the type and pitch of a musical tone to be output; and a data length of the phoneme segment from which the control section reads a phoneme segment from the phoneme memory and outputs it. A table memory that stores information regarding the number of repetitions and an output cycle, and a Dm conversion section that converts the digital musical tone signal output from the control section into an analog musical tone signal (liii
Therefore, the L table memory should be output from the control unit when synthesizing musical tones of different pitches of the same instrument using phoneme pieces required for synthesizing musical tones of a specific pitch of a certain musical instrument. A musical tone synthesis device characterized in that it is configured to refer not only to information regarding the time period of a digital musical tone signal, but also to information regarding the number of repetitions that approximates at least the vertical trienvelope of a natural musical tone.