Music – Instruments – Electrical musical tone generation
Utility Patent
1998-02-20
2001-01-02
Ro, Bentsu (Department: 2837)
Music
Instruments
Electrical musical tone generation
C084S604000, C084S622000, C084S626000
Utility Patent
active
06169241
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention generally relates to a music tone generating apparatus for generating a music tone by use of waveform data stored in a wave table memory. This music tone generating apparatus is applicable to a sound source of an electronic musical instrument, a game machine, a personal computer and so on.
2. Description of Related Art
In a typical music tone generating apparatus, waveform data is read from a wave table memory at a rate matching a pitch of a musical tone while an envelope of the read waveform data is controlled so as to generate the music tone. Such a music tone generating apparatus based on the wave table memory has limited ability of controlling timbres at waveform reproduction. A music tone may be formed by steps of preparing plural pieces of waveform data in the wave table memory, selecting the waveform data having a timbre corresponding to performance data from the prepared data, and reading the selected waveform data. For example, a waveform having characteristics corresponding to a particular performance expression is stored in the wave table memory. Actually, the performance expression varies like a short slur and a long slur, and the shape of the music tone waveform vary accordingly. It is impracticable to store all musical tone waveform variations into the wave table memory. Therefore, in order to control a timbre according to performance information, a method is generally practiced in which the waveform data read from the wave table memory is processed or modified by a digital filter having frequency characteristics corresponding to the performance information.
Anyway, the reading of waveform data is only controlled according to the pitch of a music tone to be generated. This causes a problem that the time axis of the waveform data cannot be freely controlled without regard to the pitch of the musical tone. For example, if the reading rate is increased, the pitch goes up but the whole length of the waveform is simply decreased. Conversely, if the reading rate is decreased, the pitch goes down but the whole length of a waveform is simply increased. Also, each time length of leading section, middle section, and trailing section in one waveform is determined by the pitch of the music tone.
If the time axis of the waveform data read from the wave table memory can be arbitrarily controlled, the number of timbres that can be derived from one type of waveform data can be increased. For example, different timbres could be created by altering an attack length of the music tone while maintaining the pitch. Performance expression can also be broadened significantly and diversely. For example, in the reading of a recorded slur waveform, if the waveform is compressed along time axis without altering the pitch, a slur shorter than that at recording could be created. Conversely, if the waveform is expanded, a longer slur could be generated. In the reading of a vibrato waveform, if the waveform is expanded along time axis without altering the pitch, vibrato could slow down; if the waveform is compressed, vibrato could quickens. Either way, the waveform must be expanded or compressed along time axis independently of the pitch.
In the field of voice recording/reproducing, technologies are known in which, in order to make slurred words intelligible, a voice waveform is expanded along time axis without altering the pitch. In another way, the pitch of a reproduced voice is restored to the original pitch at double-speed reproduction. It is possible to apply these technologies to the above-mentioned music tone generating apparatus. However, the pitch of music tones dynamically varies as the waveform data progresses. The above-mentioned time-axis expanding and compressing technology is only applicable to audio signals requiring no pitch control, and therefore hardly applicable to situation in which pitch control on a cent basis is required as in the sound source of an electronic musical instrument. While a music tone waveform must be controlled in different modes for different sounding operations according to the performance information, the conventional time-axis expanding and compressing technology is designed for uniformly processing all waveform data. Therefore, the conventional time-axis expanding and compressing technology involves a problem that the rate of reading waveform data cannot be freely controlled according to the pitch of a music tone to be generated.
Waveform data having a characteristic corresponding to a certain performance expression may be stored into a wave table memory. The shape of the waveform may be altered by skipping or repeating a part of this waveform at the reading from the wave table memory. In such a case, minutely observing the original waveform data, individual periods of the waveform are usually not constant. Therefore, an attempt to perform partial skip or repeat of periods contained in the waveform data simply during the reading from the wave table memory may cause poor joint at boundary, and may make difficult the waveform processing operation for joining the periods of the waveform.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to provide a music tone generating apparatus capable of freely controlling a pitch of a music tone while allowing the time-axis compression and expansion of a waveform read from a waveform memory, thereby smoothly joining plural pieces of waveform data.
In carrying out the invention and according to one aspect thereof, there is provided a music tone generating apparatus comprising: a waveform memory for storing a plurality of waveform units of one music tone in which a waveform is divided in unit of a plurality of periods to define each waveform unit, of which cycle length is normalized; first address means for generating a read address incrementing at a rate corresponding to a specified pitch of the music tone and for reading the plurality of the waveform units from the waveform memory according to the generated read address; second address means for outputting a virtual address varying temporally; and address control means for generating an alternate read address different from the above-mentioned read address by an integer multiple of the normalized cycle length according to a difference between the above-mentioned read address and the above-mentioned virtual address, and for controlling the above-mentioned first address means such that the above-mentioned plurality of the waveform units are read by the above-mentioned alternate read address instead of the original read address. Thus, this novel constitution can control compression and expansion of time axis of the music tone by the virtual address, thereby allowing the user to control as desired both of the pitch of the music tone to be generated and the compression and expansion of the time axis of the waveform data to be read from the waveform memory. This constitution also allows the user to accurately control the compression rate in the time-axis in the middle of the waveform reading operation. Since the waveform is divided in unit of a plurality of periods, the divided waveform units can be joined smoothly. In addition, the waveform unit is normalized, and the alternate read address differing from the current read address by an integer multiple of the cycle length can be generated, thereby facilitating the joining of the divided waveform units at changing of the read addresses.
In carrying out the invention and according to another aspect thereof, there is provided the music tone generating apparatus further comprising a compression rate memory for storing a compression rate to be used when the above-mentioned music tone waveform is normalized to the above-mentioned sequence of the elementary or individual waveform units. The above-mentioned first address means reads the compression rate from the compression rate memory to alter the rate of reading the waveform unit according to the compression rate. Thus, this novel constitution can not only reproduce the wavef
Fletcher Marlon T.
Morrison & Foerster
Ro Bentsu
Yamaha Corporation
LandOfFree
Sound source with free compression and expansion of voice... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Sound source with free compression and expansion of voice..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Sound source with free compression and expansion of voice... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2510451