Music – Instruments – Electrical musical tone generation
Reexamination Certificate
2002-03-14
2003-06-10
Donels, Jeffrey (Department: 2837)
Music
Instruments
Electrical musical tone generation
Reexamination Certificate
active
06576827
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a sound data transfer method, a sound data transfer apparatus, and a program thereof for generating musical sounds, voices, and any other sounds based on reading of waveform data from a relatively low-speed storage medium such as a hard disk, which has a relatively slow access rate to stored data. More particularly, the present invention relates to generation or synthesis of waveforms which faithfully express a timbre of musical sound in response to performer's various performance styles or articulation characteristic of natural musical instruments. The present invention is widely applicable to not only electronic musical instruments, but also devices, apparatuses or methods in all fields having a capability of generating musical sounds, voices or any other sounds such as mechanical instruments, computers, electronic game apparatuses and other multimedia devices. In this specification, a musical sound waveform is not limited to the pure musical sense, but may include a waveform of a voice or any sound.
There is widely known the “waveform memory reading” technology which stores waveform data (i.e., waveform sample data) encoded with any encoding system such as PCM (pulse code modulation), DPCM (differential PCM) and ADPCM (adaptive differential PCM), and which reads the stored data according to desired musical pitches to generate or synthesize a musical sound. Further, there are known various types of “waveform memory reading system” technologies. Most conventional “waveform memory reading system” technologies are used to generate a waveform of a single sound from the beginning to the end of sounding duration. One example is to store waveform data of a whole waveform of a sound from the beginning to the end of the sounding duration. Another example is to store waveform data of a partial at an attack portion of the sound with complicated changes and to store a specified loop waveform of a subsequent part at a sustain portion with little changes. In this specification, the term “loop waveform” signifies a waveform which is read repeatedly in a loop.
As means for storing waveform data, there are known ROM, RAM, hard disks, CD-ROM, etc. A hard disk and CD-ROM provide a low unit price per unit storage capacity and are suited for storing a great amount of data. However, the hard disk and CD-ROM offer a slow and unstable access speed, incapable of immediately reading necessary waveform data just at the timing of outputting a musical sound signal. For this reason, the following technologies are proposed.
Japanese Patent Unexamined Publication No. Hei. 6-308964 corresponding to the U.S. Pat. No. 5,463,183 discloses the technology of preliminarily transferring to RAM a beginning portion of a plurality of waveform data stored in a hard disk. Namely, when a sounding instruction is supplied, an operation starts advance reading of a succeeding portion of the pertinent waveform data from the hard disk, and concurrently reproducing the beginning portion of the data previously stored in RAM. After the operation completes the reproduction of the beginning portion of the data, the succeeding portion of the waveform data is reproduced.
Japanese Patent Unexamined Publication No. Sho. 63-181188 discloses the technology of reproducing waveform data by sequentially reading. Waveform data to be sequentially sounded is predefined as sequence data. This technology predetermines the time to start reading of each waveform data corresponding to a sequence timing so that the start time becomes earlier than the note-on timing.
SUMMARY OF THE INVENTION
However, the technology according to Japanese Patent Unexamined Publication No. Hei. 6-308964 needs to previously transfer beginning portions of all waveform data to RAM, thereby hindering the efficient use of the precious RAM. Since the technology according to Japanese Patent Unexamined Publication No. Sho. 63-181188 is based on the assumption that a musical sound signal is generated non-realtime, it is impossible to apply this technology to a case where music sequence data is supplied realtime.
The present invention has been made in consideration of the foregoing. It is therefore a first object of the present invention to provide a sound data transfer method, a sound data transfer apparatus, and a program thereof for efficiently and extensively using RAM and for realtime generating a musical sound signal concurrently by using a hard disk etc.
The technology according to Japanese Patent Unexamined Publication No. Hei. 6-308964 needs to previously transfer beginning portions of all waveform data to RAM. The technology according to Japanese Patent Unexamined Publication No. Sho. 63-181188 always requires a hard disk to be accessed for every event, thereby increasing a noise from the hard disk and shortening the life thereof. In addition, frequent accessing of the hard disk consumes extra resources, which should be allocated to other processes.
The present invention has been made in consideration of the foregoing. It is therefore a second object of the present invention to provide a sound data transfer method, a sound data transfer apparatus, and a program thereof for decreasing unnecessary accesses to a hard disk etc.
Conventionally, the “waveform memory reading system” technology stores waveform data of a whole waveform of one sound from the beginning to the end of the sounding duration, or stores waveform data of a part of one sound as an attack portion. Such technology must store a great number of various waveform data corresponding to various playing styles (or articulations), requiring a huge storage capacity for storing these diverse waveform data.
The aforementioned system for storing waveform data of the whole waveform can faithfully express timbre variations due to various playing styles (or articulations) specific to natural musical instruments. However, such system may simply reproduce musical sounds just in accordance with the stored waveform data, providing limited controllability and editability. For example, it is very difficult to provide characteristics control such as timing axis control of waveform data corresponding to a desired playing style (or articulation) according to the musical performance data.
The present invention has been made in consideration of the foregoing. It is therefore a third object of the present invention to provide a waveform synthesis method, a sound data transfer apparatus, and a program thereof capable of easily, simply and realtime generating high-quality waveform data available for various playing styles (or articulations).
The technology according to Japanese Patent Unexamined Publication No. Hei. 6-308964 needs to previously transfer beginning portions of all waveform data of various timbres to RAM, thereby sacrificing the RAM utilization. Further, the technology according to Japanese Patent Unexamined Publication No. Sho. 63-181188 always requires a hard disk to be accessed for every event, thereby increasing a noise from the hard disk and shortening the life thereof. In addition, frequent accessing of the hard disk scarifies limited resources of a machine.
The present invention has been made in consideration of the foregoing. It is therefore a fourth object of the present invention to provide a sound data transfer method, a sound data transfer apparatus and a program thereof for saving accesses to a hard disk etc.
In order to achieve the first object, the present invention is characterized by having the following configurations. It should be noted that examples are parenthesized.
The configuration according to a first aspect of the invention is a sound data transfer method of using a low-speed storage device (hard disk
109
) for storing sound data (vector data) of musical sound waveforms and a high-speed storage device (cache memory
44
) for caching the sound data, and transferring a part of sound data stored in the low-speed storage device to the high-speed storage device. The method is characterized by having: a timbre specification reception
Donels Jeffrey
Morrison & Foerster / LLP
LandOfFree
Music sound synthesis with waveform caching by prediction does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Music sound synthesis with waveform caching by prediction, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Music sound synthesis with waveform caching by prediction will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3110974