JPH067322B2 - Automatic adaptive accompaniment device - Google Patents

Description

The present invention relates to an automatic adaptive accompaniment device that provides accompaniment for songs, musical instruments, etc. that can be sung by anyone.

The present apparatus can be used not only for accompaniment of songs but also for accompaniment of performances of musical instruments and the like, but the case of accompaniment of songs will be described below.

Conventionally, there has been a karaoke device which provides an accompaniment of a song by reproducing an accompaniment sound recorded on a tape or the like, so-called karaoke, but it has been impossible to freely set a key which is the main tone of the song. Since singing keys vary greatly from person to person, such a karaoke device has a drawback that it is difficult for some people to sing. In addition, it is possible to automatically play an electronic musical instrument according to predetermined musical score data. This can be achieved, for example, by using the sound function of a PC-6000 personal computer manufactured by NEC. In this case, since it is an electronic musical instrument, it is possible to freely set keys. By automatically playing the accompaniment of a song using such a device, an automatic accompaniment device in which keys can be freely set can be provided.

However, setting the keys to the song required a great deal of skill and could not be done by anyone.

An object of the present invention is to provide an accompaniment that is easy for anyone to sing by automatically setting the accompaniment key according to the song. In order to achieve this object, the automatic adaptive accompaniment apparatus according to the present invention detects a pitch of a musical sound such as a singing voice or a musical instrument sound and outputs it as a time series of pitch data, and a melody composed of a plurality of note strings. , Comparing the time series of the pitch data of the musical tone and the time series of the pitch data in the melody storage section with the melody storage section that stores the time series of the pitch data corresponding to each note. It is characterized by comprising a melody matching section for obtaining a key and an automatic performance section for automatically performing an accompaniment in accordance with the key.

Examples will be described in detail below with reference to the drawings. First
The figure is a block diagram showing an automatic adaptive accompaniment apparatus according to the present invention. The singing voice input from the microphone 1 is input to the pitch detection unit 2 as a voice signal V. The pitch detector 2 detects the pitch frequency of this singing voice. There are various methods for detecting the pitch frequency of speech. For example, IEEE TRANSACTIONS ONNASSP Vol. ASSP
-24, No. 5 P399-418 "A Comparative Per
formance Study of Several Pitch Detection Algorith
ms ”. The autocorrelation method is easy to use and highly effective.

The pitch thus obtained is logarithmically converted by the logarithmic converter 3. By performing the logarithmic conversion in this way, it is possible to make the pitch difference correspond to the pitch difference on the score. The logarithmic converter 3 outputs a time series of the pitch data P. The pitch data P is held in the data buffer 4 for a certain length of time.

When the pitch data P is input for a certain length of time,
The melody matching unit 5 matches the pitch data PM in the melody storage unit 6. The melody storage unit 6 holds the melody of the sung song in the form of time series of pitch data. Matching is performed for each predetermined length of the melody.

FIG. 2 is a diagram showing an example of a musical score, and FIG.
It is a wave form diagram which shows an example of the pitch data P when a human sings the melody shown by the score in the figure. This is called an input pitch pattern. Further, FIG. 4 is a waveform diagram showing the same melody correctly with pitch data. The horizontal axis in the figure is the time axis,
The vertical axis represents the pitch height. The melody storage unit 6 stores data as shown in FIG. This data is called standard pitch pattern. Although these two patterns are matched, the tempo and pitch of the melody sung by a human are unstable.

Therefore, the time axis is first expanded and contracted to match the time direction. The start end of the input pitch pattern is matched to the standard pitch pattern of the defined time length, and the pattern pitch is expanded and contracted within a certain fixed range to obtain the pattern distance at that time. The inter-pattern distance is obtained by integrating the absolute value of the pitch difference between the two patterns at each time point. The optimum expansion / contraction amount is the expansion / contraction amount when the inter-pattern distance takes the minimum value within the expanded / contracted range. 2 at this time
The time average value of the pitch difference between the two patterns is the key difference between the standard pitch pattern and the input pitch pattern. This value is output as the key control signal K by the melody matching unit 5. The above matching operation is performed at certain intervals, and the key control signal K is output each time. The automatic performance section 7 performs an automatic performance in accordance with the key control signal K. The accompaniment sound is output from the speaker 8.

The above is one embodiment of the present invention, and the present invention is not limited to this. For example, the pitch detection method and the matching method can be considered.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, and FIG.
FIG. 3 is a diagram showing an example of a musical score, and FIG. 3 is a waveform diagram showing an example of changes in pitch data when a human sings the melody of FIG.
FIG. 4 is a waveform diagram showing an example of correct pitch data of the melody of FIG. In the figure, 1 is a microphone, 2 is a pitch detection unit, 3 is a logarithmic conversion unit,
Reference numeral 4 is a data buffer, 5 is a melody matching section, 6 is a melody storage section, 7 is an automatic performance section, and 8 is a speaker.

Claims (1)

[Claims] 1. A pitch detection unit for detecting a pitch of a musical sound such as a singing voice or a musical instrument sound and outputting it as a time series of pitch data, and a melody composed of a sequence of a plurality of notes, when a pitch data corresponding to each note is generated. A melody storage unit for storing in the form of a sequence, a melody matching unit for comparing the time series of the pitch data of the musical tone and the time series of the pitch data in the melody storage unit to obtain a key of the music, and the key An automatic adaptive accompaniment apparatus comprising: an automatic performance section for automatically performing an accompaniment according to the above.