Patent Grant
9286863
Electronic and electro-mechanical musical organs have always suffered from lack of expression because the tones produced were simply keyed on and off, could be sustained indefinitely with no attack or decay and were rather pure, unchanging tones. In wind-driven pipe organs this problem is solved with a mechanical tremulant that varies the wind pressure at a sub-audible rate imparting a vibrato, or pitch variation, and a tremolo, or volume variation, to the tones of the pipes, thus adding excitement to the sound. Often multiple tremulants were used for separate ranks of pipes.
In addition to the tremulant, another technique to improve the sound of a wind organ is the celeste, where multiple ranks of slightly off-tuned pipes where played together to produce a very low frequency undulation in the sound and a complex cancellation and reinforcement of harmonics. This effect is very similar to the process that makes stars twinkle, or move around and vary in brightness. Different paths through the atmosphere have slightly different delays and cause the light to cancel or add and appear to arrive from a slightly different spot in the heavens. The wave lengths of sound waves are much longer than visible light, so the effects of cancellation and reflection can move the sound around a larger space in a pleasing and exciting way.
In electric organs, the vibrato effect is often imparted electronically; though, this is less than ideal as the sound is too precise and comes from a single speaker. If the celeste effect is attempted by purely electronics means, it suffers the same short-coming of emanating from a single or pair of speakers. In live performance venues, only the listeners in the stereo sweet-spot hear the spatial effects while most listeners, including the musician, miss out. In the case of the pipe organ, the pipes are physically spread; and the sounds come from multiple directions. To achieve both the vibrato and spatial effect for electric organs, it became common to use a mechanically-orbited speaker where the sound sprays in different directions as the sound transducer spins about. Many attempts have been made to capture the sound of the mechanically-orbited speaker by electronic means and reproduced by a stereo-sound system with disappointing results.
As the transducer orbits in a mechanically-orbited speaker; the apparent source of sound, the mouth of the horn, moves toward and away from the listener. As the source moves toward the listener, the pitch would rise; and as the sound moved away, the pitch would fall. This pitch change is due to the Doppler Effect. The sound would reflect from various surfaces in the room, producing the spatial effect. In a happy accident, the cabinet selected for the original commercial models of mechanically-orbited speakers interacted with the orbiting source of sound and added a celeste-like effect. At low speeds the celeste effect dominates over the Doppler Effect and is a highly desired feature of the speaker. Unfortunately, this cabinet construction also limits the sound output of the speaker due to the narrow slots. In efforts to increase the sound output, modifications to the cabinet lost the celeste effect.
While mechanically-orbited speakers have been popular in the past, they suffer from several drawbacks. To produce the vibrato effect over the desired range of musical frequencies, the transducer must spin. The speaker cabinet must be rather large and heavy, making it difficult to transport to live shows. The mechanical parts are delicate, requiring frequent maintenance. Attempts have been made to implement mechanically-orbited speakers with rotary joints to conduct sound signals to the orbiting transducers, but noise from the sliding contact and maintenance issues caused this approach to be abandoned.
Synchronizing multiple mechanically-orbited speakers is difficult, and a single physically rotating transducer has limited sound volume output. Venues have grown in size; and audiences have come to expect a full sound, so many performers resort to placing the orbiting speaker in a sound-isolated location, using a microphone and sound amplification system with multiple speakers. This results in loss of the desirable interaction of the speaker and the listening room.
Because the physical size of the orbiting speaker defines the acoustic performance, smaller and less expensive mechanically-orbited speakers do not achieve the desired musical effect, especially losing the desired frequency modulation by only rotating instead of orbiting.
Often mechanically-orbited speakers have only two speeds and no opportunity to vary the effect without physically modifying the speaker, thus, having a very limited expressiveness. This leads to playing techniques that resort to rapidly switching the drive motors on and off in an attempt to achieve intermediate speeds.
Keyboard players often have two or more instruments, or even one instrument, that can emulate more than one acoustic instrument, such as a synthesizer, that can produce tonewheel organ or piano sounds. Mechanically-orbited speakers have difficulty reproducing piano sounds without coloring and cannot reproduce uncolored piano and orbiting organ sound simultaneously. The traditional solution to this problem has been to add an additional pair of amplifiers and speakers dedicated to the stationary channels, adding weight, size and cost to the speaker system.
There is a need in the art for a speaker system to amplify musical instruments that can produce the desired vibrato, tremolo and celeste spatial effects while being lighter and more rugged for transport, having no moving parts to avoid frequent maintenance, being able to achieve the desired vibrato, tremolo and celeste spatial effects in a low-cost configuration, or being able to be driven at high power levels and having multiple orbiting speakers ganged for higher sound levels, having the ability to vary the musical effect for increased expressiveness, and producing uncolored sound simultaneously with vibrato, tremolo and celeste spatial effects.
It is, therefore, the object of the present invention to enable a sound-system design for live performance of music with realistic tremulant and celeste spatial effects in a physical configuration that lends itself to light-weight portability, low maintenance and low-cost manufacture. Instead of mechanically-orbited sound transducers, the present invention uses two or more fixed sound transducers facing in different directions with the sound signal modulated separately for each transducer, as to impart the sense of orbiting as the sound reflects around the room and of a very low frequency undulation in the sound and a complex cancellation and reinforcement of harmonics due to combination of sound from a plurality of differentially directed transducers.
The sense of direction of sound in the present invention is increased by using fixed-sound transducers or groups of transducers selected and arranged to produce the appropriate sound radiation pattern for the desired effect. The transducer array extends the tremulant and celeste effects to lower frequencies and selectively deepens the effects over that achieved by existing mechanically-orbited speakers.
Because the orbited effect is imposed electronically, the same set of amplifiers and sound transducers can be simultaneously used to amplify and project sound without the tremulant and celeste or other effects or with a different set of effects. This makes it practical for a musician to use a single sound system for organ-sound reproduction with strong tremulant at the same time as electric piano having no tremulant, or with a light tremulant appropriate for the desired piano sound. This multiuse can be extended to any combination of instruments, voice or any other sound source.
Embodiments of the present invention lend themselves to addition of other sound effects particular to orbiting speakers, including simulation of horn-throat distortion, overdrive of the amplifier, amplifier- and speaker-cabinet emulation, and spatially-diverse reverberation simulation. For various reasons described below, mechanically-orbited speakers have limitations on the amount of sound output a single unit can produce and ganging multiple units may result in the tremulant effect being degraded. The present invention achieves higher sound level from multiple transducers in a single unit and allows for multiple amplifier/transducer units to be ganged for even higher sound levels while maintaining the quality of the synchronized tremulant and celeste effects.
FIG. 1—depicts a rank of wind driven organ pipes with a tremulant.
FIG. 2—shows the cross section of a pipe organ tremulant.
FIG. 3—depicts a rank of pipe organ pipes with tremulant and celeste.
FIG. 4—shows the internal mechanism of a popular organ sound system.
FIG. 5—shows the outside view of the sound system of
FIG. 6—shows the sound paths from the transducer in the sound system of
FIG. 7—is a cross section of the horn used in the sound system of
FIG. 8—is an amplitude/frequency plot of a comb filter.
FIG. 9—is a block diagram of an electronic flanger.
FIG. 10—is a block diagram of a stereo electronic flanger.
FIG. 11—depicts the stereo sweet-spot.
FIG. 12—depicts the full coverage of an electronically-orbited speaker system.
FIG. 13—shows an electronically-orbited speaker enclosure where the cone-type transducers have a front-loaded horn.
FIG. 14—shows the construction details of the electronically-orbited speaker.
FIG. 15—shows the basic signal connections of an electronically-orbited speaker.
FIG. 16—is a plot of the four amplitude envelopes of the four signals intended for the four sets of sound transducers, one signal for each face of the cabinet, in this case a complex modulation.
FIG. 17—is a signal flow diagram of a fully featured signal processor.
FIG. 18—is a block diagram an embodiment of an electronic celeste, the subject of the present invention.
FIG. 19—is a top view of the sound paths from the transducers of an electronic celeste to the listener's ears.
FIG. 20—is a block diagram a second and reduced complexity embodiment of an electronic celeste.
FIG. 21—is a block diagram single delay buffer embodiment of an electronic celeste.
The detailed description set forth below in connection with the appended drawings is intended as a description of exemplary embodiments of the present invention and is not intended to represent the only embodiments in which the present invention can be practiced. The term “exemplary” used throughout this description means “serving as an example, instance or illustration” and should not necessarily be construed as preferred or advantageous over other exemplary embodiments.
This detailed description includes specific details for the purpose of providing a thorough understanding of the exemplary embodiments of the invention. It will be apparent to those skilled in the art that the exemplary embodiments of the invention may be practiced without these specific details. In some instances, well-known structures and devices are shown in block diagram form in order to avoid obscuring the novelty of the exemplary embodiments presented herein.
In particular, the exemplary embodiment is described in terms of a unit with four faces with associated amplifiers and sound transducers; but this number could be any number from two to some larger number limited by cost and complexity concerns. The faces of the unit may be deployed horizontally, vertically or in any configuration where the sound of two or more transducers is directed in different directions.
In the exemplary embodiment, the signal processing function is described as, but not limited to, a digital signal processor. Some or all of the signal processing may be implemented by other means, such as, but not limited to, analog circuits, standard computing components, or any other electronic or electrical means.
The term “celeste” is used herein to mean a sound effect that imparts a sense of motion and a variation in the relative strengths of the fundamental and harmonic tones in a periodic way. This variation usually takes place at a slow rate of a few times per second or once every few seconds.
The term “orbiting speaker” is used herein to mean any form of loudspeaker or sound-amplification device or sound-modification device intended to modulate or change the sound of a musical instrument or other sound source in a periodic way, especially to impart a periodic varying of pitch, amplitude or spatial perception. Orbiting speaker is intended to cover speakers that are commonly referred to as rotating or rotary speakers. The orbiting speaker may be a stand-alone device or collection of devices, or be part of a musical instrument.
The term “rotate” is used herein to mean the motion of an object that turns on its own axis.
The term “orbiting” is used herein to mean the circular motion of an object around a central point. In particular, the horn mechanism in a mechanically-orbited speaker rotates. The apparent sound source at the mouth of the horn orbits.
The terms “signal processor”, “digital signal processor” or “DSP” may be a purpose designed computing device, a general purpose computing device, a collection of analog or digital electronic circuits, or a combination of any of the above.
The term “transducer” is used herein to mean a device that converts one form of energy into another. In particular, a sound transducer converts electrical signals into sound waves.
The term delay is used herein to mean a time delay that causes a sound to arrive at a later time to the listener. The term delay, when applied to celeste effect, means a time delay of less than 5 milliseconds, which is perceived by a listener to be a single sound when heard along with the original sound. The term delay, when applied to reverberation effect, means a time delay in excess of 5 milliseconds, which is perceived as a separate sound when heard along with the original sound.
A major part of the thrilling sound of a full wind-driven pipe organ is the undulating sound imparted by the tremulant mechanism 100 depicted schematically in
A cross section of a pipe organ tremulant is shown in
A rank of organ pipes with a celeste is shown in
When two pipes sound the same or make a very close musical note, the sound takes on a richness and spatial character. Pairs of pipes are often tuned slightly off to produce a beat note at a very low frequency between one-half to 3 beats per second. Because the musical notes are coming from different positions, the waves from each pipe sometimes reinforce the other and sometimes cancel. The harmonics reinforce and cancel at different times and rates, derived from the very low frequency beat notes, producing a varying timber and depth not achievable with a single rank of pipes or with pipes of different types played together. This slow variation in harmonic content and sense of source of the sound of a musical note is highly sought after by musicians. A more recent method of producing the sound of a tremulant and celeste uses a sound source that moves in a circle or orbit.
Existing orbiting-speaker effect units are available in two forms: mechanical or electronic. The mechanism of the mechanical-type orbiting speaker is depicted in
The rotary horn 401 is counter-balanced by a dummy horn 402. Sound is produced by a compression-driver unit 404 and is passed upward through a rotary joint and pulley 403. The horn assembly is rotated by an electric motor and belt 405. In most modern orbiting speakers, there is a second motor, not shown, to rotate the horn at a different speed, providing both a high- and a low-speed modulation effect.
The low-frequency, cone-type speaker 410 fires downward into a rotating drum 411 made of light-weight wood or other material. The drum has a scoop-shaped section that turns the sound toward the side of the cabinet 400 where there are slots to allow the sound to exit the cabinet. The drum 411 is rotated by the electric motor and pulley system 412. Like the horn in modern units, there are two motors, one for high speed and one for low speed. The second motor and clutching system is left out of the diagram for simplicity. The drum 411 is typically rotated in the opposite direction than that of the horn 401. Because of the limitations of the size of the low-frequency sound transducer 410 and rotary drum, there is little frequency modulation effect; and the amplitude modulation is imparted largely by the mouth of the deflector drum passing by the slots in the cabinet.
As the playing style of many tonewheel organists includes switching often between the high speed and low speed or even stopping rotation of the speakers, the belts and clutching mechanisms 403, 405, 412 require frequent maintenance.
How the mechanically-orbited speaker produces the celeste effect is depicted in
Because these sound ports 602 are constricted in size, the sound coming out each port is close to the same intensity, which is key to producing the desired celeste effect. As shown in
A popular explanation of the operation of a mechanically-orbited speaker ascribes the tremulant sound to the Doppler Effect. While this may be true, it is only part of the story. The celeste effect described here becomes dominant at the slower speed. Experiments show that rotating the horn very slowly by hand reveals sharp changes in the tone quality produced by phase cancellation, something quite different and richer than the simple frequency modulation produced by Doppler alone.
In the orbiting speaker this process of reinforcement or cancellation varies with frequency. The fundamental of a tone may be reinforced while a harmonic is cancelled or vice versa. As the horn turns in the cabinet, the pattern of reinforcement and cancellation verses frequency changes as the sound path lengths change, as represented by curves 801, 802, 803. This means fundamental tones will be changing independently of harmonics as the horn revolves. Thus, the designer of the mechanically-orbited speaker captured the sound of a true celeste by imparting physical motion to the sound. The combination of the orbiting of the omnidirectional sound source (the horn with diffuser) and the constricted cabinet ports creates the spatially separated sound sources with time delays.
In the quest to produce a celeste effect, many electronic means have been used. Analog bucket-brigade devices and digital-delay lines are popular. All of these means are represented by the block diagram 900 in
As the delay is varied the comb filter depicted in
In an attempt to provide some spatial movement, the flanger may be implemented in a stereo pair as shown in block diagram 1000 form in
In
Some mechanically-orbited speaker manufacturers have not understood the value of the interaction of the orbiting sound transducer interacting with the cabinet. They produce products with stereo microphones mounted inside the cabinet in an attempt to capture the tremolo and celeste effect so the sound of the orbiting speaker may be reinforced with a larger sound amplification system. The result is a strong amplitude modulation effect with limited frequency modulation and very little of the celeste effect.
The solution to the needs enumerated above is the electronically-orbited speaker of the present invention described herein. The US application published as US2013/0163787 describes the basic operation of the electronically-orbited speaker system and is incorporated herein in its entirety.
A simple exemplary embodiment would consist of four separate acoustic transducers mounted on the four vertical faces of a box. Each transducer would be driven by a separate electronic amplifier. The amplitude of drive to each transducer is modulated by any electronic means. Improvements in high-power, class D audio amplifiers and switch-mode power supplies make it very practical and cost effective to have several amplifiers integrated into one product. These amplifiers and power supplies are very efficient, produce little heat and are considerably lighter in weight than traditional audio amplifiers.
In
Between the two conditions described above, when the virtual orbiting-sound source is pointing at the corner of the box, the drive levels to both the front and right side transducers 1301, 1305, 1303 are equal and at some typically lower-power level, thus sounding like a single transducer pointed toward the corner of the box. This process continues, handing off the audio power from one transducer to the next as the virtual-sound source orbits in a complete circle. This is a very simple description of the action of the electronically orbited speaker; and there are many factors that improve the musical effect, which are described in more detail in the referenced application and more briefly below.
Fundamental to the success of the orbiting-speaker effect is the radiation patterns of the acoustic transducers. The patterns must be narrow enough to provide the desired effect of the sound spraying out in different directions as the virtual-sound source orbits. If the transducers have a very wide radiation pattern, there is little change in the sound no matter which way the speaker is pointed. The ideal transducer pattern for an electronically-orbited speaker with four sides would be a single beam approximately 90° wide. However, real transducers do not have a single radiation pattern for all frequencies. Higher frequencies tend to have a very narrow pattern; while at lower frequencies, the pattern broadens until it becomes almost omnidirectional. This effect is dominated by the size of the transducer. A transducer with an effective diameter of approximately one wavelength produces the ideal 90° pattern.
wavelength=344/f Sound wavelength formula
f in Hertz
wavelength in meters
The exemplary embodiment 1300 shows an improvement to the sound radiation pattern of the mid-range frequencies, extending the orbiting sound-source effect to lower frequencies in a simple but effective manner. Considering a single cone-type transducer 1301 of a typical 150 millimeter diameter, and using the formula, the transducer loses directionality below 2293 Hertz. By adding a second identical transducer 1302 spaced 0.75 meter center-to-center, a line array is established. The effective diameter for the horizontal radiation pattern calculation becomes 0.75 meter, considerably narrowing the pattern at lower frequencies. The proper directional radiation pattern would be maintained down to 458 Hertz, thus extending the orbiting sound-source effect into the middle of the musical spectrum in a simple and cost-effective manner. The scheme is extended to each face with the transducers 1303 and 1304 making up the line array of the right face.
Classical mechanically-orbited speakers often have a deflector plate attached to the mouth of the horn in an attempt to spread the sound radiation pattern at higher frequencies. Musicians often removed the deflector plate and/or removed parts of the cabinet to achieve different musical sounds. In the electronically-orbited speaker system, the DSP may divide a frequency band dedicated for a specific transducer into sub-bands and modulate the signal with different amplitude envelopes for each sub-band to equalize the sound radiation pattern between sub-bands. Alternatively, the amplitude envelopes may be selected to emphasize the difference of radiation pattern between sub-bands. By doing so under player control, the electronically-orbited speaker system can emulate the sound of different models and configurations or modifications of classical orbiting speakers.
Some players prefer to use two classical mechanically-orbited speakers. Each rotor in each speaker rotates at a slightly different rate, making for a very complex variation in the tremulant effect. The DSP of the electronically-orbited speaker system may use a plurality of amplitude envelopes running at different speeds to provide this complex tremulant effect. The delays to impose a reverberation effect may be different for each amplitude envelope to provide the illusion that the virtual rotors are in different physical locations.
Mechanically-orbited speakers often have a feature where the rotor is stopped by a brake with the transducer facing front to provide maximum sound level in the non-orbiting configuration. The DSP of the electronically-orbited, upon receiving a command to stop the virtual orbiting, may continue the current orbit until the virtual transducer reaches front center where it may stop and ramp the amplitude to maximum regardless of the amplitude envelope at that point. As the virtual orbiting transducer accelerates or decelerates, the amplitude envelope may be changed to emphasize the tremulant effect.
The size of a transducer also affects the efficiency in converting the electrical signal to sound. A smaller transducer works better at higher frequencies while a larger transducer is needed to reproduce the longer wavelengths of lower frequencies. The electrical signals may be divided into bands appropriate for each transducer by a crossover network and each transducer driven with only the signals it can reproduce well. Dividing the musical spectrum up to be reproduced by separate transducers also helps the variation in radiation pattern with frequency. A smaller transducer maintains a narrow radiation pattern only in the upper frequencies. A larger transducer provides a narrow radiation pattern at lower frequencies, though even a very large transducer becomes omnidirectional at the lowest musical frequencies.
This problem of needing to use different sizes of transducers is not unique to electronically-orbited speakers. The best mechanical-orbited speakers use a rotating high-frequency transducer and a separate rotating low-frequency transducer. To deepen the musical effect, the two transducers are typically rotated in opposite directions.
The electronically-orbited speaker in another exemplary embodiment uses a high-frequency transducer and one or more low-frequency transducers on each of the four faces of the box. Each transducer or transducer array has an associated amplifier. The DSP divides the input signal into two frequency bands performing the crossover function, one band for the high-frequency transducers and one band for the low-frequency transducers. The DSP then divides the signal for each band into four signal streams for each of the transducers on each face of the box. Each of the two sets of four streams is coupled to the associated amplifier. The orbiting is imparted by the same amplitude envelope method as described above. In this case, the high-frequency set is orbited in one direction and the low-frequency set is optionally orbited in the opposite direction.
In further description of
The sound transducers 1301, 1302 are behind a front-loaded horn made up of the transducer baffle, the walls of the cabinet and the septum 1310. This arrangement allows the center of sound radiation of each transducer to be close to the edge of the cabinet for widest speaker array possible and provides some improvement in efficiency of the transducer.
Class D power amplifiers can achieve high power with efficiency exceeding 90 percent, which reduces heatsink size and ventilation air requirements. Class D amplifiers of medium-power are available quite economically in a single integrated circuit package, while high-power amplifiers may be constructed with minimal component size and number. A switch-mode power supply eliminates the large and heavy 50/60 Hz power transformer and also operates at high efficiency. This makes it practical to have multiple amplifiers, one for each horn-type transducer and one for each pair of cone-type transducers. For extremely high-power applications, the electronics may be mounted external to the transducer cabinets, driven by cooling considerations. The signal processor function may be integrated into the speaker cabinet, part of a separate electronics package that includes the power amplifiers, or may be a separate physical unit or connected units. The signal processor function may be performed by a standard computing device coupled with the amplifiers and sound transducers. The computing device may be a personal computer, laptop computer, notebook computer, tablet computer, smartphone or any other computing device.
A line-level audio signal is provided to the Instrument Input 1510 by the sound source, typically a musical instrument. The signal is processed by the DSP 1520 by being split into four pairs of signals used to drive each face of the unit and modulated to produce the orbiting effect. Four signals are bandpass filtered by the DSP for the mid frequencies and four signals highpass filtered for the high frequencies. Pairs of signals, one mid and one high, drive each face. Each face has a dedicated power amplifier 1531 to drive the pair of cone-type transducers 1502 and a second power amplifier 1530 to drive the high frequency horn transducer 1501. A line-level subwoofer output 1515 is filtered by the DSP to only allow the bass frequency signal to pass to the subwoofer (not shown) for amplification by a larger transducer.
The PA Input 1511 is a stereo pair. It is similarly, but separately, split and processed by the DSP and summed at each of the nine outputs to the amplifiers. The right and left channels of the stereo pair are summed into the faces of the speaker with different gains to produce the desired stereo effect.
The control input 1512 may be of one or more types and multiple types may be incorporated into any particular embodiment. One interface may emulate the existing popular mechanically-orbited speaker with discrete signals for fast and slow speeds and a brake to stop rotation. Other control inputs may use the Musical Instrument Digital Interface (MIDI) protocol to control various parameters of operation including, but not limited to, fast and slow speed, stop, variations in speed for each of the high- and mid-frequency channels, acceleration and deceleration of the virtual rotors, crossover frequencies, envelope profile selection (described later), distortion effects thresholds and many other parameters. The MIDI interface may use the MIDI signaling definition or be implemented via Universal Serial Bus (USB) as are many musical instruments.
The fundamental orbiting speaker effect is produced by splitting the input signal into one path for each face of the speaker enclosure and amplitude modulating the paths separately to sweep the sound in a complete circle. The sound is physically moved by imposing the appropriate amplitude envelope on the signal paths. The process of physically moving the apparent sound source and direction in a circle imposes both amplitude and frequency modulation on the sound. The amplitude modulation can be depicted as amplitude envelopes as shown in
This process is repeated for each orbit of the sound. The slow-speed effect is called “Chorale” and typically orbits at 45 RPM. The fast-speed effect is called “Tremolo” and orbits at approximately 400 RPM. Some models of mechanically-orbited speakers have multiple pulleys to change the orbiting speed but require disassembly of the cabinet to make the change. In the electronically-orbited speaker several orbital speeds are available, changed via the external control input. Similarly, the external control input provides selection of multiple amplitude envelopes for different orbiting effects.
To orbit in the opposite direction, the count is decremented instead of incremented. To produce these amplitude envelopes, a look-up table may be used or the values of the envelope calculated in real time. In the case of the look-up table, the full rotation may be represented by the segment of one of the envelopes from step 0 to step 64. The remainder of the circle may be simply produced by modulo arithmetic and incrementing the look-up table pointer up or down for the positive or negative slope of the curve. The shape of the envelope may be asymmetric to emulate an off-center rotating horn. The look-up table implementation is ideal for producing arbitrary envelope shapes. The speed of stepping through the envelope look-up table may be varied to emulate the variations, such as loose or worn belts and/or pulleys, inherent in mechanically-orbited speakers.
The physical configuration of sound transducers is key to producing the orbiting speaker sound, but the electronics provide an increased level of control and variation desired for a flexible sound-reproduction system. Though the effects described here may be implemented in various technologies, the DSP is powerful and cost effective. This exemplary embodiment will be described in terms of a DSP with embedded software that implements the components of the signal path.
The signal is then split into frequency bands by the highpass 1715, bandpass 1716 and lowpass 1717 filters. The signal from the highpass 1715 filter feeds the tweeters, the bandpass 1716 feeds the mid-range transducers and the lowpass filter 1717 is routed to a line-level output to be connected to the subwoofer. The signals from the PA or alternate instrument inputs 1720, 1725 routed through a similar set of signal function blocks, though the parameters, such as distortion models and crossover frequencies, may be different from the instrument input 1710 channel. There is a reverberation emulation block 1730 for each signal path. By splitting the reverberation emulation across the faces of the speaker, a spatial aspect of the reverberation effect is introduced that is lacking in the typical front-facing, stereo-sound system. The signals for the left, right and rear channels have different delays and arrive at the listener from different directions due to room reflections. This better emulates the effect of a larger room. Also, the tweeter and mid-range signal paths are treated differently to add frequency dependent aspects to the effect.
In this embodiment, the signal from the instrument input is shown as being routed through the amplitude envelope processing 1735, though the PA channel 1720, 1725 may optionally be routed through the instrument envelope processing or through a separate set of envelope processors (not shown).
The signals output from the envelope processors are then fed to signal mixers and then on to the appropriate amplifiers and sound transducers as shown in
Starting at the PA inputs, one for the left 1720 and one for the right 1725 stereo pair, there is a compressor/limiter function 1721 and crossover filters 1722, 1723 and 1724 as described previously. The crossover frequencies for the PA channel may be different from the instrument channel. In particular, the crossover frequencies for the instrument-signal path are selected to provide a narrow sound-radiation pattern as described in the support for
The lowpass 1724 signals from both the left and right PA channels are routed to the subwoofer mixer 1745, because the lowest frequencies have no apparent directional characteristics. The highpass 1722 and mid-range 1723 signals from both the left and right PA channels are routed separately to the mixers 1740. These channels would have further signal processing blocks (not shown) for amplifier and speaker cabinet effects, reverberation and amplitude envelope.
Before the PA channels are mixed, there may be individual gain adjustments to place the stereo image properly by routing the signals to the appropriate face or faces of the speaker system. In a simple example, the left PA-tweeter channel from the highpass filter 1722 would be routed with full gain at gain adjustment 1741 to the left-tweeter output mixer; and the right-channel highpass signal would be routed with full gain through gain adjustment 1743 to the right-tweeter output mixer. The left-channel mid-range signal from the crossover filter 1723 would be routed with full gain at the gain adjustment 1742, and the right-channel signal routed with full gain at gain adjustment 1744 to the right mid-range output mixer. In this simple example, all other gain adjustments would be set to the lowest setting to block the signals. For situations where a broader coverage was desired, the signals at lower gain may be routed to the front and rear faces. This is especially useful in cases where the audience surrounds the player.
The electronic celeste 1800 is comprised of an input 1801, envelope generator 1807, variable delay lines 1810, 1811, 1812, 1813, amplifiers 1820, 1821, 1822, 1823 and sound transducers 1830, 1831, 1832, 1833.
The variable delay lines 1810, 1811, 1812, 1813 in one example are implemented as a circular buffer in DSP memory. The analog signal is typically sampled at a rate at least twice the highest frequency in the signal to be processed to meet Nyquist Criterion. A common sampling rate used is 48 Kilo samples per second with 24 bits of precision. The samples are stored in an area of random access memory coupled with the DSP. The DSP writes new samples into the buffer as they are converted from analog. An address pointer is used to keep track of where the last sample was stored. A second address pointer is arranged to point at a sample that was previously stored. Every time a new sample is stored, both pointers are incremented. Modulo arithmetic is used when incrementing the pointers such that the area of memory is used over and over, appearing to be a never ending circle. The second address pointer, besides being incremented every time a new sample is stored, may also be adjusted to vary the delay between the sample being stored in the buffer and the sample being retrieved from the buffer. This delay may be from zero to a few milliseconds.
To avoid a sound artifact known as zipper noise, caused by discontinuities in the sound waveform from changing the delay, the variable delay line may be oversampled, making the effective delay difference between delay buffer addresses smaller. To oversample, the input sample may have zero value samples inserted between each real sample, and then the result is low-pass filtered to create interpolated intermediate samples. The delay buffer must be made larger by the oversampling factor. An alternative technique is to use a fractional-delay line whereby the output address pointer has a fractional part that is used to interpolate between a pair of samples from the buffer. The oversampling and fractional-delay line techniques may be combined to further reduce zipper noise.
Assuming a sampling rate of 48 Ksps, to implement a delay line with a maximum delay of 3 milliseconds would require that the buffer length be at least 144 samples. Using the speed of sound of 340 meters per second, this delay represents a distance of 1.02 meters, the width of the largest orbital speaker cabinet. Delays larger than 5 milliseconds start to become heard as a reverberation instead of a celeste effect, so longer delays should be avoided. A delay of 3 milliseconds represents a phase shift of 180 degrees at 166.6 Hertz, the lowest frequency where a celeste effect may be reasonably created. These dimensions and performance match closely with those of mechanically-orbited speakers commonly in use.
A signal from a musical instrument or other source is introduced at the input 1801. The signal is divided four ways and couples to four variable-delay lines 1810, 1811, 1812, 1813. The output of each variable-delay line is coupled with the input of the associated amplifier 1820, 1821, 1822, 1823. The output of each amplifier is coupled to one or more sound transducers 1830, 1831, 1832, 1833, each pointed in a different direction. The amplitude of each signal path is modulated to impose the orbiting effect as described above and not shown for clarity. The delay of each signal path is varied, controlled by envelope generator 1807. The envelope generator produces a separate envelope for each delay line. The envelopes may be identical in shape but offset in phase from each other. For a four-sided cabinet, the phases would typically be offset by 90 degrees. The envelope generator operates in a fashion similar to the envelope generator used to drive the amplitude modulation as depicted in
The envelope generator for orbiting and the envelope generator for celeste may or may not be operated in synchronization, allowing further variation and richness to be added to the sound of the instrument. The delay envelope may also be asymmetric to emulate an off-center rotation of a horn. The delay envelope may be as simple as a straight line, resulting in a triangle waveform that causes the output address pointers to sweep back and forth across the delay buffer at a constant rate. The delay envelope may be complex, as depicted in
Unlike the mechanically-orbited speaker where the length and depth of delay that produces the celeste effect are fixed in the physical configuration of the rotating transducers and the cabinet, in the electronically-orbited speaker, the celeste effect is implemented in a signal processor that may be programmed to operate with various parameters under control of the musician. The delay envelope may be generated by incrementing and/or decrementing the output address pointers to produce increasing and decreasing delays. Different delay envelopes may be selected by addressing a different look-up table or applying mathematical functions to a single look-up table. Delay envelopes may be generated by mathematical functions or any other method that may generate periodic or aperiodic waveforms. A user interface may be implemented as part of the electronically-orbited speaker to command the signal processor to change any or all of the parameters described above. The parameters of the electronic celeste and the parameters of the electronically-orbited speaker may be controlled by an external device, such as a computer or other musical instrument. The external control may be communicated via a music industry standard interface, such as MIDI or MIDI over USB, or by any other communications medium.
The sounds via the two paths sometimes add constructively and sometimes destructively producing the comb-filter effect depicted in
A signal from a musical instrument or other source is introduced to the input 2101 where it is compressed, sampled, filtered, modulated to inducing orbiting and optionally other effects not shown here for clarity. The samples representing the sound signal are coupled to the variable delay line 2110 where the samples are stored in memory; and an input address pointer locates the samples while the address pointers increment with modulo arithmetic, as described in
As can be seen from the description above, the electronically-orbited speaker with electronic celeste produces the rich and exciting sound of classical orbiting speakers while achieving many advantages, such as ease of transport, upgradability, higher sound-level output, easily changed characteristics and operation with two or more sound systems with independent characteristics.
Those of skill would appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the exemplary embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps above have been described generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the exemplary embodiments of the invention.
The various illustrative logical blocks, modules, and circuits described in connection with the exemplary embodiments disclosed herein may be implemented or performed with a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, analog circuit, discrete hardware components, vacuum tube, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.
The steps of a method or algorithm described in connection with the exemplary embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), flash memory, Read Only Memory (ROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), registers, hard disk, a removable disk, a CD, DVD, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in an electronically orbited speaker with electronic celeste. In the alternative, the processor and the storage medium may reside as discrete components in an electronically-orbited speaker with electronic celeste.
In one or more exemplary embodiments, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media, including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EPROM, EEPROM, Flash, CD, DVD or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. Disk and disc, as used herein, includes compact disc (CD), optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers or LEDs. Combinations of the above should also be included within the scope of computer-readable media.
The previous description of the disclosed exemplary embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these exemplary embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the sprit or scope of the invention. Thus, the present invention is not intended to be limited to the exemplary embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
| Number | Name | Date | Kind |
|---|---|---|---|
| 3535431 | Szabo | Oct 1970 | A |
| 3644657 | Miller | Feb 1972 | A |
| 3749837 | Doughty | Jul 1973 | A |
| 3809792 | Deutsch | May 1974 | A |
| 4000676 | Love | Jan 1977 | A |
| 4072079 | Sharp | Feb 1978 | A |
| 4162372 | Gross | Jul 1979 | A |
| 4280388 | White | Jul 1981 | A |
| 4308422 | Schmoll, III | Dec 1981 | A |
| 4308428 | Finch | Dec 1981 | A |
| 4329902 | Love | May 1982 | A |
| 4342247 | Ippolito et al. | Aug 1982 | A |
| 4392405 | Franz et al. | Jul 1983 | A |
| 4428270 | Ippolito | Jan 1984 | A |
| 4622878 | Sharp | Nov 1986 | A |
| 5763807 | Clynes | Jun 1998 | A |
| 5848166 | Fisher et al. | Dec 1998 | A |
| 5998722 | Kondo | Dec 1999 | A |
| 6873708 | Levy | Mar 2005 | B1 |
| 6931134 | Waller, Jr. | Aug 2005 | B1 |
| 20130163787 | Moon | Jun 2013 | A1 |
| 20150078584 | Moon | Mar 2015 | A1 |
| Number | Date | Country | |
|---|---|---|---|
| 20150071451 A1 | Mar 2015 | US |
