Apparatus And Method For Simulating Chiff In A Sampled Amplitude Electronic Organ

Abstract

A sampled amplitude electronic organ embodies first and second memories storing groups of amplitude samples that, respectively, delineate energy levels of the wave shape of a musical tone, and energy levels of the wave shape having the dominant energy of a third harmonic or its equivalent of the musical tone. The groups of amplitude samples are read out at rates determined according to actuation of a number of keyboard switches so that the repetition rate of the group as read from the memory is the chosen frequency of the musical tone to be produced by operation of an actuated keyboard switch. The two groups of amplitude samples are added for about the first seven cycles of read out to thereby introduce a transient chiffing wave to the initial portion of the primary tone to be generated.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to methods and apparatus for enhancing tonal qualities of an electronic organ and more particularly concerns the simulation of chiff in an electronic organ embodying sampled amplitude signal generation.

2. Description of Prior Art

The term "chiff" as used in todays organ design refers to a well-known phenomenon in wind-driven organ pipes. A common defect of flue or labial organ pipes is that when they are first blown, they tend to sound at a higher harmonic than the fundamental pitch they produce after the initial transient. Shortly after the initial transient, the pitch quickly drops to the normal steady state frequency and tone. This phenomenon of the dominant higher harmonic that occurs during the initial transient has been called by many names, including the name "chiff". In fact, certain pipes have been given the name "spitz flote" because of the analogy to a "spit type tone during the initial transient.

Chiffed tones were long considered to be undesirable and a defect in the organ pipes. During the last century organ voicers have developed several techniques which eliminate or greatly reduce chiffing. These techniques usually involve introduction of mechanical knicks along the organ pipes languid (languet). However, as subjective evaluation of music and tonal qualities evolves, todays pipe organs and electronic organs have undergone a retrogressed form of innovation. Changing musical preferences favor the building of new organs which have certain of the tonal advantages and also the defects of the Baroque period. Following this reversion to older musical tonal characteristics, organs are now deliberately voiced to have at least several ranks of pipes that chiff. In particular the 8 foot voice is desirably chiffed in present day organs.

In several electronic organs that are commercially available, chiff is simulated by introducing a short grace note during the attack time of a tone that is to be chiffed. The grace note is most commonly selected as the nearest equal tempered note to the third harmonic of the fundamental tone being keyed. It has been judged that such third harmonic provides an optimum simulation of a chiff organ pipe. However in such an arrangement the chiffing note is mechanically coupled with the primary note being played, or at best electronically coupled for a time predetermined according to conventional delay circuits. In such an arrangement, the duration of the chiffing transient will have little or no relation to the frequency of the note being played. Further, prior systems are not applicable to sampled amplitude organs.

Accordingly it is an object of the present invention to provide a method and apparatus for simulating chiff in a sampled amplitude musical tone generation system.

SUMMARY OF THE INVENTION

In carrying out principles of the present invention in accordance with a preferred embodiment thereof, groups of representations of primary and chiff amplitude samples collectively delineating the wave shape of a musical tone and a wave shape having a predetermined relation to the first wave shape are generated at group repetition rates that determine frequency of the primary musical tone to be played. The groups of chiff amplitude sample representations are combined with the groups of primary amplitude sample representations during an initial transient period. A further feature of the invention is control of the duration of the initial transient period in accordance with a number of cycles of the primary group repetition rate.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1a represents a wave form of a musical note having an amplitude envelope showing an initial attack and a final decay,

FIG. 1b illustrates a chiffing wave form,

FIG. 1c illustrates the desired combination of the normal wave form and the chiffing wave form,

FIG. 2 is a block diagram of a sampled amplitude musical tone generation system embodying chiffing simulation of the present invention.

DETAILED DESCRIPTION

As illustrated in FIG. 1, a normal wave form of a musical tone takes the general shape shown in this figure. Amplitude of the tone builds up from its time of its initiation, t.sub.0, during the attack which terminates at a time t.sub.1. During the major portion of the tone, from time t.sub.1 until time t.sub.2 the amplitude envelope remains substantially constant. At time t.sub.2 a decay period is initiated which terminates with termination of the note at time t.sub.3.

A desirable chiffing wave form is shown in FIG. 1b, which illustrates a tone having its dominant energy content at the third harmonic of the tone of FIG. 1a. This chiffing wave form also builds up in its amplitude envelope during the attack time, from t.sub.0 to a point just preceding the end of the attack time t.sub.1. At a point at the end of the attack time, or slightly before, amplitude of the chiffing wave form rapidly falls to zero.

Combining the wave shapes of FIG. 1a and FIG. 1b yields the wave shape shown in FIG. 1c. This is the desired chiff wave form, having a dominant third harmonic during the attack interval t.sub.0 through t.sub.1. At about time t.sub.1 the chiffing wave form rapidly falls to zero and the primary wave form continues as in the normal wave form.

Apparatus for generating the wave forms of FIGS. 1a and 1b and combining these to provide a chiffed wave form is embodied in the stored sampled amplitude organ that is illustrated in block form in FIG. 2. The system illustrated in FIG. 2 is basically the same as the systems shown in U. S. Pat. No. 3,610,799, for Multiplexing System for Selection of Notes and Voices In An Electronic Musical Instrument and U. S. Pat. No. 3,639,913, for Method and Apparatus for Addressing a Memory at Selectively Controlled Rates, both issued to George A. Watson, and assigned to North American Rockwell Corporation, the assignee of the present invention. Although the invention is described herein as applied to the digital organ system of the Watson patent, it will be readily appreciated that it is not limited to systems employing digital representations of sampled amplitudes. Principles of the invention may also be applied to instruments and systems that delineate complex wave shapes of musical tones by means of other types of representations of sampled amplitudes. Such other representations may include various well known forms of analog arrangements, such as voltage, current, electrical charge, and the like.

Briefly the organ that is described in full detail in the Watson patent and generally illustrated in FIG. 2 hereof, embodies a multiplexor 24 that provides a series of output signals on a line 25, each of which occurs in a unique specifically allocated time slot of each multiplexor cycle. As the operator actuates a given key or pedal or some combination of keys or pedals of the instrument, the arrangement scans each key and pedal during each multiplexor cycle and produces a pulse or no pulse at particular time slot allocated to a given key, depending upon whether such key or pedal has been actuated. The multiplexed signal on line 25 is fed to a generator assignment logic circuit 26 which feeds the pulses representing actuated keys or pedals to a tone generating circuit including a phase angle number selector 28, a phase angle register 32, a sample point address register 34, and a gate 30. Although in an actual system there may be as many as twelve such tone generating circuits, only one is shown herein and only one is necessary for an understanding of the practice of the present invention. Where several of such tone generators are employed, each may be identical to the others. The function of the generator assignment logic is to direct a signal from the multiplexor that represents actuation of a given key to one of the tone generators that is not already engaged in receiving a signal and producing a tone therefrom.

The phase angle number selector 28 which may be common to all of a number of tone generators and shared by these, selects (either from storage or by repetitive calculation) a number from a set of distinct and different numbers that vary according to the twelfth root of two. As is well known, a semi tone, or half tone in the musical scale of equal temperament is the frequency ratio between any two tones whose frequency ratio is the twelfth root of two. Therefore, the several numbers of the set calculated by or stored in selector 28 identify phase angles or frequencies of the individual notes of the scale of notes to be played. These phase angle numbers identify read out rate of stored sampled points of the complex wave form for the respective note frequencies in the entire range of frequencies of the musical scale of the particular instrument. Details of such calculation and/or storage together with circuitry therefore are set forth in the above-identified U.S. Pat. 3,639,913.

The tone generator, by means of its sample point address register 34, addresses a memory 40 by means of an address decoder 42. When generator assignment logic 26 determines that a particular tone generator is claimed (available for reception of the next note identified in the multiplexed signal), gate 30 is opened to allow a number corresponding to a particular note or actuated key that is to be assigned to this tone generator to be introduced into the phase angle register 32.

Phase angle register 32 feeds the number stored therein to the sample point address register 34 and, upon each clock pulse received from a sampling clock, augments the number stored in the sample point address register 34 by the number in the phase angle register. As the sample point address register is augmented, its count advances and is fed to address decoder 42 which then addresses the memory 40. Stored in the memory 40 are groups of representations of amplitude samples that collectively delineate the wave shape of a musical tone. These groups of representations, which are eight bit digital words in the exemplary embodiment, are read out of the memory at a group repetition rate that determines the frequency of the musical tone. The memory output is fed to circuitry 29 for further processing and conversion to a musical tone. This circuitry may accumulate various signals, impose attack and decay amplitude envelopes, combine various voices, and convert the resulting electrical signal to an audible tone, all as more particularly described in the above-identified U.S. Pat. 3,610,799.

Thus it will be seen that the arrangement described to this point, all of which is further detailed in the above identified Watson patent, responds to actuation of a selected key on the instrument keyboard, selects a phase angle number according to the identity of the actuated key, gates the selected number into the phase angle register, and then causes the sample point address register to be repetitively augmented by the number stored in the phase angle register at a rate determined by a sampling clock. As the sample point address register is augmented, the number contained therein is decoded to address the memory 40. The memory thereupon provides on output lines generally identified at 44, groups of representations of amplitude samples that collectively delineate the wave form of a primary note that has been selected by the particular key actuated.

According to principles of the present invention, a chiffing arrangement is added to the above-identified system of Watson to produce a resulting wave form such as illustrated in FIG. 1c. To this end, there is provided a second memory 50 which may comprise a separate section of the primary memory 40. Memory or memory section 50 has its own address decoder 52 which also decodes the same address number contained in sample point address register 34. Therefore, as sample point address register 34 is augmented to achieve a step by step addressing of the various sample representations in memory 40, it also achieves a step by step addressing of the amplitude sample representations in memory 50. Preferably the amplitude sample representations in memory 50 represent the energy content or amplitudes, at selected points along the time axis, of a complex wave form having a dominant third harmonic of the wave form represented by the amplitude samples stored in memory 40.

For example, where amplitude sample representations stored in memory 40 represent an eight foot voice, the chiffing wave form is preferably stored in the form of amplitude sample representations for an eight foot voice having a dominant third harmonic. However, it will be readily appreciated that, as an alternative, the chiffing wave form may be stored as amplitude sample representations of a sixteen foot voice having its energy concentrated in its sixth harmonic. For storing representations of a sixteen foot voice, twice as many memory representations are located in the memory as in the case of an eight foot voice. Accordingly for a given rate of addressing the memory, it takes twice as long to scan the sixteen foot voice whereby a lower frequency is produced.

As the sample point address register is augmented, primary representations of amplitude samples are read out of memory 40 and chiff amplitude sample representations are read out of memory 50. These amplitude sample representations, which are digital words in the described embodiment, are combined in an adding network 54 which then feeds the circuitry 29 to impose attack, decay, shaping, voicing and the like, and convert the selected and combined signals to an audio tone.

As previously mentioned the chiffing signal to be combined with the primary signal is only a transient. Accordingly, the chiffing signal is read from memory 50 only for a short duration of the initial portion of the read out. In the embodiment illustrated in the drawings, the read out of the chiffing wave form from memory 50 is controlled by an inhibiting gate 60 that is normally open to allow read out. The gate closes to block further read out upon receipt of an inhibit signal from a chiff counter 62. Chiff counter 62 receives a counting pulse input on line 64 from the sample point address register 34. A counting pulse occurs on line 64 once during each full cycle of address register 34. A preferred arrangement is to employ a stage of the address register 34 that is an order of higher or highest significance so that such stage is uniquely logical one when the last memory location is being addressed. In other words, chiff counter 62 will advance one count for each read out cycle of the memory, and will advance one count for each group of amplitude sample representations that is read from the memory.

Initiation of the counting of the chiff counter 62 is enabled by a key depress signal appearing on line 66 from the generator assignment logic 26 as more particularly described in the above-identified Watson patent. The key depress signal indicates initiation of a note. When the chiff counter 62 attains a predetermined count, such as the count of seven for example, it feeds an inhibit signal to close gate 60 and thereby block any further read out from the chiff memory 50. It will be readily appreciated that the location of gate 60, as shown in FIG. 2, is exemplary only. Alternatively this gate may be located at the output of the memory. Further, the inhibit signal may be otherwise applied, so as to terminate the combining of chiff and primary amplitude sample representations at the desired moment. For example, upon occurrence of the predetermined count of chiff counter 62, the primary amplitude sample representations on lines 44 may be caused to bypass adder 54 so that chiff amplitude representations are no longer combined therewith.

Although the chiff counter 62 is shown as a separate counter, in order to more readily describe its functions, this counter may simply be comprised of a group of higher order stages of sample point address register 34.

It will be seen that a significant advantage of the described system derives from the fact that the transient duration of the chiffing signal, as effectively combined with the primary signal, varies according to the frequency of the output or desired primary signal. In the described arrangement regardless of the frequency of the tone to be played and regardless of the repetition rate of the groups of amplitude representations read from the memories, the chiff counter will inhibit addition of the chiff amplitude signals after the selected number of initial cycles of the desired wave form. This timing of the chiff transient is readily selected and precisely controlled.

The foregoing detailed description is to be clearly understood as given by way of illustration and example only, the spirit and scope of this invention being limited solely by the appended claims.

Claims

I claim:

1. The method of simulating chiff in an amplitude sample musical tone generation system comprising the steps of

generating repetitive groups of primary amplitude sample representations collectively delineating the wave shape of a musical tone, said groups being generated at a group repetition rate that determines frequency of the musical tone,

generating repetitive groups of chiff amplitude sample representations collectively delineating the wave shape having a dominant harmonic of the fundamental of the tone to be generated,

combining said groups of chiff amplitude sample representations with said groups of primary amplitude sample representations only during an initial transient period of the generation of said groups of primary amplitude sample representations, and

producing an electrical signal representing said primary amplitude sample representations as modified by transient combination with said chiff amplitude sample representations.

2. The method of claim 1 wherein said step of adding is initiated substantially simultaneously with initiation of generation of said groups of amplitude sample representations.

3. The method of claim 1 wherein said step of combining said chiff and primary amplitude sample representations is caused to terminate at the end of an interval that varies in accordance with the frequency of the tone to be generated.

4. The method of claim 1 wherein said step of combining said chiff and primary amplitude sample representations comprises the steps of storing said chiff amplitude sample representations, reading said chiff amplitude sample representations from storage, and inhibiting said reading of chiff amplitude sample representations when the combining of said primary and chiff amplitude signal representations is to be terminated.

5. The method of claim 1 wherein said step of combining chiff and primary amplitude sample representations comprises the steps of counting groups of said primary amplitude sample representations and inhibiting said combining when a predetermined number of said groups of primary representations has been counted.

6. An electrical musical instrument comprising

a first memory for storing a first group of amplitude sample representations delineating energy levels of a complex wave form,

a second memory storing a second group of amplitude sample representations delineating energy levels of a wave shape having a dominant harmonic of said complex wave form,

memory addressing means including a sample point address register for simultaneously and repetitively addressing both said first and second memories to cyclically read out therefrom both said first and second groups of amplitude sample representations at said repetitive address rate,

means for selectively inhibiting read out of said second memory,

a cycle counter connected to count cycles of said sample point address register,

means for initiating counting of said cycle counter,

means responsive to attainment of a preselected count by said cycle counter for actuating said inhibit means to inhibit read out from said second memory,

means for adding first and second groups of amplitude sample representations read out of said first and second memories, and

means responsive to said adding means for producing a combined output signal including an initial third harmonic transient thereon.