JP2961998B2 - Sound source device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an analysis / synthesis sound source which analyzes and stores musical tone waveform data for each spectrum and re-synthesizes them to form a musical tone signal. In particular, the present invention facilitates interpolation between partial data. Related to the synthesized and synthesized sound source.

[0002]

2. Description of the Related Art Various tone signal forming methods for electronic musical instruments have been conventionally proposed. The main one is FM
There are a system sound source and a waveform memory system sound source. The FM system sound source is a system in which a plurality of basic signals are synthesized (frequency-modulated) by various algorithms and intensities to form musical sound waveform data.
The waveform memory system is a system in which musical sounds of natural musical instruments are stored in a memory in time series, and when sound is generated, these are sequentially read out to form musical sound waveform data.

[0003]

In the FM system tone generator, the waveform can be controlled flexibly and flexibly by adjusting the parameters. However, it is difficult to control the waveform to the intended tone and the reproducibility of a natural musical instrument is poor. Was. The sound source of the waveform memory system has good reproducibility of musical tones of natural musical instruments, but it is difficult to process the waveform, so that it is not possible to change the timbre.

Further, when there are a plurality of tone data, the natural tone waveform has a fluctuation especially in an attack part and the frequency (spectrum) moves. However, when a plurality of tone sounds are interpolated to form an intermediate tone color, Requires interpolation for each spectrum. However, this is not possible with any of the conventional FM system sound source and waveform memory system sound source.

SUMMARY OF THE INVENTION An object of the present invention is to provide a sound source device having good reproducibility of the tone color of a natural musical instrument and good controllability of a waveform, and to enable smooth interpolation between tone colors in the tone source device. It is in.

[0006]

According to the first aspect of the present invention, a plurality of bandpass filters for dividing and covering a tone band are provided .
Channels that are adjacent to each other
The cutoff part of the filtration zone of
Noto, enter the musical tone waveform data for each channel, each channel
Frequency data of the filtered spectrum output from the channel ,
And analyzing means for detecting the intensity data, frequency data and intensity data for each channel said analyzing means has analysis Ji
Comprising storage means for storing as Yan'nerudeta, and a synthesizing means for synthesizing a tone waveform data reading each channel data from said storage means, said analyzing means, adjacent
From both channels
Identical spectrum with over-band frequencies output
Time, based on the intensity data detected from both channels.
The intensity data of one channel data
And the intensity data of the other channel data is set to 0
Characterized in that it. The invention according to claim 2, wherein the memory device
In the row, each channel data is sorted in the order of the sound waveform data.
Stores a plurality of data as the sound waveform data progresses
Frequency data of the spectrum moving between channels
And intensity data with the channel data of the same channel.
And storing it.

[0007]

According to the sound source apparatus of the present invention, a plurality of musical tone waveform data
Is analyzed for each channel (bandpass filter), and stored as channel data including frequency data and intensity data. The channel re-synthesis of the musical sound waveform data
This is done by synthesizing the data . Since the channel data is extracted from the musical sound waveform data, data with good reproducibility can be created using sampling sounds of natural musical instruments. The channel data of the natural musical instrument San
Since the data is related to the spectrum of the pulling sound or the like, the workability is good and the timbre can be easily changed.
Furthermore, one spectrum spans multiple channels .
If this is detected, this is
By managing it as a
Even when the spectrum moves between a plurality of channels in time , the overtone relationship is clarified and the interpolation is facilitated by managing the channel data of the same channel in time series .

[0008]

FIG. 1 is a block diagram of a basic circuit of an analysis and synthesis system sound source to which the present invention is applied.

This sound source has an analysis unit 1, a storage unit 2, an interpolation unit 3, a shift unit 4, and a synthesis unit 5. FIG. 2 shows the configuration of the analysis unit 1. The analysis unit 1 has a configuration in which a plurality (128) of band-pass filters (BPFs) are connected in parallel. Each BPF is identified by a channel number of channels 0 to 127. The bandwidth of each BPF is 125 Hz, and its filtration band is 0 channel: 0 to 125 H
z, one channel: 125 to 250 Hz, three channels: 250 to 375 Hz, from 0 Hz to 16 k
Hz. In addition, each channel
As shown in FIGS. 2 and 7,
Burlap. Further, the BPF has an FFT function, and can detect frequency data and intensity data Mag of a spectrum passing through the BPF.
The frequency is detected as a deviation from the band center frequency. By adding the deviation to the band center frequency, frequency data Freq of an absolute value is obtained. The analysis unit 1 can be composed of a BPF bank as described above,
It can also be realized using an FT analyzer. Furthermore, the number of channels is not limited to 128, and the bandwidth is 125 Hz.
However, the present invention is not limited to this.

The PCM waveform data is input to the analyzer 1. The PCM waveform data is sampled with a sampling clock of 32 kHz. The analysis unit 1 handles continuous 2048 samples as one set of data (one frame). However, each frame is output every 64 samples, and adjacent frames overlap with a shift of 64 samples (see FIG. 3). Analysis unit 1
When one frame of sampling data is input, each channel outputs frequency data Freq and intensity data Mag of the frame. Here, the number of samples forming one frame is not limited to 2048, and the frame interval is not limited to 64 samples.

[0011] The storage unit 2, the frequency data F ^m _n and intensity data M ^m _n of each channel are stored as a table for each frame (Fig. 4). A group of tables of a plurality of frames corresponding to one waveform data is called voice data. This voice data is stored for each tone. In the case of multi-sampling (a method of storing a plurality of waveforms of one timbre for each tone range or intensity), one timbre includes a plurality of voice data.

The analysis of the waveform data (converted into voice data) by the analysis unit 1 and the storage in the storage unit 2 are preprocessing performed before the performance.

The interpolation section 3 is a circuit section for forming data of each sampling clock when reading voice data and forming waveform data. That is, although the frame data is output every 64 clocks, 6 frames between each frame are output.
Frequency data and intensity data at three sampling timings are calculated by linear interpolation from the frame data before and after that (see FIG. 5). The interpolation is performed using the same channel data for each channel. The calculated data is output to the shift unit 4.

The shift section 4 is a circuit for shifting only frequency data in order to generate a musical tone having a pitch (frequency) designated by a keyboard or the like. The shift amount is determined based on the frequency of the sampled waveform data and the designated frequency.

The synthesizing section 5 is a circuit for synthesizing frequency data and intensity data for each channel to synthesize one waveform data. The synthesis may use an inverse FFT synthesis or an addition synthesis.

The voice data stored in the storage section 2 is read out, interpolated and shifted for each channel, and then synthesized to form musical tone waveform data.
Therefore, the operations of the storage unit 2, the interpolation unit 3, the shift unit 4, and the synthesis unit 5 are real-time operations during performance.

The analysis / synthesis sound source of the above-described method can form an electronic sound by sampling an actual sound such as a natural musical instrument, and can greatly process the sampled waveform data.

Hereinafter, the operation of the analyzer 1 in the present invention will be described in detail.

FIG. 6 is a flowchart showing the processing procedure of the analysis unit 1, and FIGS. 7, 8 and 9 are diagrams for explaining a data allocation method. In the flowchart of FIG. 6, first, original PCM data is input to the analysis unit 1 (n1). This PCM data is processed into analysis parameters consisting of frequency data and intensity data (n
2, n3). Since the PCM data has a variety of positive and negative spectral intensities due to sampling or the like, the intensity data of the analysis parameter is converted into an absolute value (n4). here,
The PCM data includes a plurality of fundamental waves and their overtone spectra, each of which is included in the filter band of the BPF of any channel. Some of these spectra are distributed over two channels (see FIG. 7A). Such a spectrum gives rise to analytical parameters in both channels (FIG. 7 (B),
(C)) Since a complicated calculation is required at the time of interpolation, this is distributed to one of the channels. The distribution is performed for the channel whose intensity is large (n
5: FIG. 7 (D), (E): FIG. 9 (A), (B)). That is,
The intensity data of the analysis parameter of the channel with low intensity is added to the intensity data of the analysis parameter of the channel with high intensity. Thereafter, a component that changes the channel with time is detected (n6). For example, FIG.
T _-1, N channel before and after t _0, if at t ₁ is spectrum slide into moved from N-channel to N + 1 channel as shown in FIG. (C) t ₀ as, N + 1
The same frequency data is detected for both channels.
On the other hand, in the intensity data, the intensity is rapidly changed around t ₀ as shown in FIG. In such a case, when the states of the frequency data and the intensity data are detected, it can be determined that the spectrum is shifted. Based on this, it is desirable to process the same spectrum on the same channel even if there is pitch fluctuation in an attack part or the like, so that it is allocated on the same channel (n7). That is, in the case of FIG. 8, when the spectrum moves to the (N + 1) channel, the N band filtering band and the (N + 1) channel filtering band are exchanged so that the spectrum passes through the N channel.

Thereafter, the channel number to which the harmonic spectrum is to be assigned is determined based on the frequency of the fundamental tone. This channel is a channel to which harmonics belong when the tone waveform data is in a steady state. The overtone spectrum is shifted to this channel (n8: FIG. 9 (D)).
The above data is stored in the storage unit 2 (n9).

When a new analysis parameter is generated by interpolating with another analysis parameter (n11), an analysis parameter without noise is generated by interpolating the analysis parameters so that the harmonic relations correspond to each other. (N12). The analysis parameters are added and synthesized (n
13) New tone waveform data (PCM data) can be synthesized (n14). In order to obtain musical tone waveform data of an arbitrary frequency, the interpolation section 4
May be recombined after shifting to a predetermined frequency.

[0022]

[0023]

As described above, according to the present invention, the channel obtained by analyzing the spectrum of the sampled tone waveform data is analyzed.
By synthesizing musical tones based on the channel data, the musical tones of natural musical instruments can be reproduced well.
By storing as channel data, it is possible to realize a sound source device with high timbre processability. Also, straddle multiple channels and move between channels
Also, the one always the same channel spectrum tea
By managing it as channel data , interpolation of musical tones can be facilitated .

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of an analysis and synthesis system sound source to which the present invention is applied.

FIG. 2 is a diagram showing a configuration of an analysis unit of the analysis and synthesis system sound source.

FIG. 3 is a diagram showing an example of PCM waveform data input to the analysis and synthesis system sound source.

FIG. 4 is a diagram showing storage contents of a storage unit of the analysis and synthesis system sound source.

FIG. 5 is a view for explaining the operation of the interpolation unit of the analysis / synthesis system sound source.

FIG. 6 is a diagram showing an analysis / synthesis procedure executed in the analysis / synthesis system sound source.

FIG. 7 is a diagram showing an example of a spectrum of musical tone waveform data to be analyzed;

FIG. 8 is a diagram showing an example of a spectrum of musical tone waveform data to be analyzed;

FIG. 9 is a diagram showing correspondence between spectra and BPF channels.

[Explanation of symbols]

1-analysis unit, 2-storage unit, 3-interpolation unit, 4-shift unit,
5- Synthesis part.

Claims (2)

(57) [Claims] 1. A plurality of channels comprising a plurality of bandpass filters for dividing and covering a musical tone band ,
The cutoff of the filtration band of the adjacent channel is over.
And those are Rappu, enter the musical tone waveform data for each channel, each channel
Frequency data filtration spectrum output from the analysis means for detecting the intensity data, storage means for storing frequency data and intensity data for each channel in which the analysis means is analyzed as channel data, each channel data from said storage means And synthesizing means for synthesizing the musical tone waveform data by reading the data from the adjacent channels.
Identical switch having the frequency of the wrapping filter band
Detected from both channels when the spectrum is output
Of one channel data based on the
Determine the intensity data and the other channel data
A sound source device that sets the intensity data of the data to 0 . 2. The storage means stores each channel data.
In addition to storing a plurality of musical sound waveform data in the
A switch that moves between channels as the sound waveform data progresses
Spectrum frequency data and intensity data in the same channel
2. The storage device according to claim 1, wherein the channel data is stored as channel data of the channel.
Sound source device.