Knowledge

Karplus–Strong string synthesis

Source 📝

36: 441:, which was used to model acoustic waves in strings, tubes, and membranes. The first set of extensions and generalizations of the Karplus-Strong Algorithm, typically known as the Extended Karplus-Strong (EKS) Algorithm, was presented in a paper in 1982 at the International Computer Music Conference in Venice, Italy, and published in more detail in 1983 in Computer Music Journal in an article entitled "Extensions of the Karplus Strong Plucked String Algorithm," by David A. Jaffe and Julius O. Smith, and in Smith's PhD/EE dissertation. 366: 161: 734: 552: 753: 490:
The patent was licensed first to Mattel Electronics, which failed as a company before any product using the algorithm was developed, then to a startup company founded by some of the laid-off Mattel executives. They never got sufficient funding to finish development, and so never brought a product to
216:
loop. The filter can be a first-order lowpass filter (as pictured). In the original algorithm, the filter consisted of averaging two adjacent samples, a particularly simple filter that can be implemented without a multiplier, requiring only shift and add operations. The filter characteristics are
494:
While they may not adhere strictly to the algorithm, many hardware components for modular systems have been commercially produced that invoke the basic principles of Karplus-Strong Synthesis: using an inverted, scaled control system for very small time values in a filtered delay line to create
491:
market either. Eventually Yamaha licensed the patent, as part of the Sondius package of patents from Stanford. It is unknown whether any hardware using the algorithm was ever sold, though many software implementations (which did not pay any license fees to the inventors) have been released.
436:
recognized that the transfer-function of the KS, when viewed as a digital filter, coincided with that of a vibrating string, with the filter in the feedback loop representing the total string losses over one period. He later derived the KS algorithm as a special case of
345:
may be sharpened or flattened relative to the fundamental frequency. The original algorithm used equal weighting on two adjacent samples, as this can be achieved without multiplication hardware, allowing extremely cheap implementations.
495:
playable notes in the Western Tempered tuning system, controlled with volt per octave tracking or MIDI data. The Inventors were not specifically credited, though the term "Karplus-Strong Synthesis" is referenced in some of the manuals.
429:
conjectured that the Karplus-Strong (KS) algorithm was in some sense a vibrating string simulation, and they worked on showing that it solved the wave equation for the vibrating string, but this was not completed.
384: 498:
Hardware components capable of Karplus-Strong style synthesis include the Moog Clusterflux 108M, Mutable Instruments Elements and Rings, 4ms Company Dual Looping Delay, 2HP Pluck, Make Noise Mimeophon,
413:
Holding the period (= length of the delay line) constant produces vibrations similar to those of a string or bell. Increasing the period sharply after the transient input produces drum-like sounds.
382: 675: 233:(specifically, the lowest nonzero resonant frequency) of the resulting signal is the lowest frequency at which the unwrapped phase response of the delay and filter in cascade is 325:
filters may be used, but FIR have the advantage that transients are suppressed if the fractional delay is changed over time. The most elementary fractional delay is the
383: 257: 359:
file. The algorithm used a loop gain of 0.98 with increasingly attenuating first order lowpass filters. The pitch of the note was A2, or 220 Hz.
352:
analysis can be used to get the pitches and decay times of the harmonics more precisely, as explained in the 1983 paper that introduced the algorithm.
137: 835: 444: 422: 144:
did the first analysis of how it worked. Together they developed software and hardware implementations of the algorithm, including a custom
17: 566: 828: 772: 79: 57: 50: 1033: 856: 821: 171:
A short excitation waveform (of length L samples) is generated. In the original algorithm, this was a burst of
305:
The length of any digital delay line is a whole-number multiple of the sampling period. In order to obtain a
561: 317:
are used with parameters selected to obtain an appropriate phase delay at the fundamental frequency. Either
933: 900: 96: 1002: 946: 861: 438: 131: 99:
that loops a short waveform through a filtered delay line to simulate the sound of a hammered or plucked
793: 588: 951: 941: 866: 575: 467:, and scored for eight guitars, mezzo-soprano and computer-generated stereo tape, with a text based on 375: 310: 744: 725: 318: 788: 44: 749:
Independently controlled wavetable-modification instrument and method for generating musical sound
923: 322: 896: 683: 643: 448: 127: 61: 716:
Techniques for Digital Filter Design and System Identification, with Application to the Violin
876: 641:
Karplus, Kevin; Strong, Alex (1983). "Digital Synthesis of Plucked String and Drum Timbres".
230: 220:
The filtered output is simultaneously mixed into the output and fed back into the delay line.
111: 104: 236: 1012: 886: 613: 326: 8: 994: 984: 908: 475:. Jaffe continued to explore the musical and technical possibilities of the algorithm in 431: 971: 913: 881: 798: 700: 660: 512: 500: 306: 209: 451:-modification method for plucked-string synthesis, but only published it as a patent. 1007: 956: 871: 768: 213: 198: 100: 918: 692: 652: 479:, for computer-generated plucked strings (1982), as well as in later works such as 979: 748: 729: 464: 355:
A demonstration of the Karplus-Strong algorithm can be heard in the following
1027: 803: 556: 468: 426: 314: 187: 141: 115: 212:
of the filter must be less than 1 at all frequencies, to maintain a stable
205: 160: 844: 730:
Wavetable Modification Instrument and Method for Generating Musical Sound
349: 260: 191: 172: 149: 123: 119: 813: 704: 664: 405: 180: 696: 656: 503:, Non Linear Circuits Is Carp Lust Wrong?, and the Strymon Starlab. 555: This article incorporates text from a publication now in the 342: 217:
crucial in determining the harmonic structure of the decaying tone.
176: 434: 126:
analysis. However, it can also be viewed as the simplest class of
134:, because the delay line acts to store one period of the signal. 356: 197:
This excitation is output and simultaneously fed back into a
183: 676:"Extensions of the Karplus-Strong Plucked String Algorithm" 145: 148:
chip. They named the algorithm "Digitar" synthesis, as a
808: 459:
The first musical use of the algorithm was in the work
239: 309:
often needed for fine tuning the string below JND (
743: 728:, Alexander R. Strong, Kevin J. Karplus, " 724: 421:Due to its plucked-string sound in certain modes, 251: 341:(5)). If the phase delay varies with frequency, 110:At first glance, this technique can be viewed as 1025: 809:David A. Jaffe's music, including sound examples 530: 416: 204:The output of the delay line is fed through a 829: 640: 595: 536: 546: 544: 186:or frequency sweep, or a single cycle of a 836: 822: 674:Jaffe, David A.; Smith, Julius O. (1983). 673: 601: 843: 567:Dictionary of Greek and Roman Antiquities 80:Learn how and when to remove this message 541: 43:This article includes a list of general 454: 14: 1026: 130:-modification algorithms now known as 27:Method of physical modelling synthesis 817: 804:A HTML5 port of the above application 767:. Upper Saddle River: Prentice-Hall. 762: 713: 560: 273:is therefore calculated according to 799:More sound examples under CC license 224: 29: 487:for female chorus and tape (1987). 24: 364: 266:for a given fundamental frequency 159: 49:it lacks sufficient corresponding 25: 1045: 782: 483:for string quartet and tape, and 461:May All Your Children Be Acrobats 747:, Alexander R. Strong, " 614:"inventor:(Alexander R. Strong)" 550: 404:Problems playing this file? See 380: 34: 481:Telegram to the President, 1984 155: 93:Karplus–Strong string synthesis 718:(PhD/EE). Stanford University. 606: 175:, but it can also include any 13: 1: 518: 789:The Karplus-Strong Algorithm 417:Refinements to the algorithm 140:invented the algorithm, and 97:physical modelling synthesis 7: 751:", published 1986-11-18 732:", published 1987-03-17 506: 439:digital waveguide synthesis 329:between two samples (e.g., 302:is the sampling frequency. 132:digital waveguide synthesis 10: 1050: 765:Elements of Computer Music 763:Moore, F. Richard (1990). 745:US application 4622877 726:US application 4649783 311:Just Noticeable Difference 993: 970: 932: 895: 852: 714:Smith, Julius O. (1983). 537:Karplus & Strong 1983 477:Silicon Valley Breakdown 179:signal, such as a rapid 18:Karplus–Strong algorithm 691:(2). MIT Press: 56–69. 651:(2). MIT Press: 43–55. 64:more precise citations. 684:Computer Music Journal 644:Computer Music Journal 602:Jaffe & Smith 1983 570:. London: John Murray. 369: 253: 252:{\displaystyle -2\pi } 164: 152:for "digital guitar". 1034:Sound synthesis types 962:Karplus–Strong string 447:developed a superior 368: 315:interpolating filters 254: 231:fundamental frequency 163: 118:similar to that of a 112:subtractive synthesis 1013:Software synthesizer 857:Frequency modulation 455:Musical applications 327:linear interpolation 237: 995:Digital synthesizer 463:written in 1981 by 432:Julius O. Smith III 972:Analog synthesizer 934:Physical modelling 513:Digital delay line 501:Arturia MicroFreak 370: 249: 165: 1021: 1020: 1008:Scanned synthesis 947:Digital waveguide 862:Linear arithmetic 583:Missing or empty 576:cite encyclopedia 385: 376:Karplus-Strong #1 225:Tuning the string 214:positive feedback 103:or some types of 90: 89: 82: 16:(Redirected from 1041: 942:Banded waveguide 867:Phase distortion 838: 831: 824: 815: 814: 778: 757: 756: 752: 738: 737: 733: 719: 708: 680: 668: 628: 627: 625: 624: 610: 604: 599: 593: 592: 586: 581: 579: 571: 554: 553: 548: 539: 534: 387: 386: 367: 307:fractional delay 259:. The required 258: 256: 255: 250: 138:Alexander Strong 85: 78: 74: 71: 65: 60:this article by 51:inline citations 38: 37: 30: 21: 1049: 1048: 1044: 1043: 1042: 1040: 1039: 1038: 1024: 1023: 1022: 1017: 1003:Analog modeling 989: 980:Graphical sound 966: 928: 891: 848: 845:Sound synthesis 842: 785: 775: 754: 735: 697:10.2307/3680063 678: 657:10.2307/3680062 632: 631: 622: 620: 612: 611: 607: 600: 596: 584: 582: 573: 572: 551: 549: 542: 535: 531: 521: 509: 473:The People, Yes 457: 419: 411: 410: 402: 400: 399: 398: 397: 395: 388: 381: 378: 371: 365: 301: 292: 285: 272: 238: 235: 234: 227: 201:L samples long. 167: 158: 95:is a method of 86: 75: 69: 66: 56:Please help to 55: 39: 35: 28: 23: 22: 15: 12: 11: 5: 1047: 1037: 1036: 1019: 1018: 1016: 1015: 1010: 1005: 999: 997: 991: 990: 988: 987: 982: 976: 974: 968: 967: 965: 964: 959: 954: 952:Direct digital 949: 944: 938: 936: 930: 929: 927: 926: 921: 916: 911: 905: 903: 893: 892: 890: 889: 884: 879: 874: 869: 864: 859: 853: 850: 849: 841: 840: 833: 826: 818: 812: 811: 806: 801: 796: 794:Sound Examples 791: 784: 783:External links 781: 780: 779: 773: 759: 758: 740: 739: 721: 720: 710: 709: 670: 669: 637: 636: 630: 629: 618:Google Patents 605: 594: 564:, ed. (1870). 562:Smith, William 540: 528: 527: 526: 525: 520: 517: 516: 515: 508: 505: 465:David A. Jaffe 456: 453: 418: 415: 401: 393: 389: 379: 374: 373: 372: 363: 362: 361: 297: 290: 281: 270: 248: 245: 242: 226: 223: 222: 221: 218: 202: 195: 157: 154: 88: 87: 42: 40: 33: 26: 9: 6: 4: 3: 2: 1046: 1035: 1032: 1031: 1029: 1014: 1011: 1009: 1006: 1004: 1001: 1000: 998: 996: 992: 986: 983: 981: 978: 977: 975: 973: 969: 963: 960: 958: 955: 953: 950: 948: 945: 943: 940: 939: 937: 935: 931: 925: 924:Concatenative 922: 920: 917: 915: 912: 910: 907: 906: 904: 902: 898: 894: 888: 885: 883: 880: 878: 875: 873: 870: 868: 865: 863: 860: 858: 855: 854: 851: 846: 839: 834: 832: 827: 825: 820: 819: 816: 810: 807: 805: 802: 800: 797: 795: 792: 790: 787: 786: 776: 774:0-13-252552-6 770: 766: 761: 760: 750: 746: 742: 741: 731: 727: 723: 722: 717: 712: 711: 706: 702: 698: 694: 690: 686: 685: 677: 672: 671: 666: 662: 658: 654: 650: 646: 645: 639: 638: 634: 633: 619: 615: 609: 603: 598: 590: 577: 569: 568: 563: 558: 557:public domain 547: 545: 538: 533: 529: 523: 522: 514: 511: 510: 504: 502: 496: 492: 488: 486: 482: 478: 474: 470: 469:Carl Sandburg 466: 462: 452: 450: 446: 442: 440: 435: 433: 428: 427:Kevin Karplus 424: 414: 409: 407: 392: 377: 360: 358: 353: 351: 347: 344: 340: 336: 332: 328: 324: 320: 316: 312: 308: 303: 300: 296: 289: 284: 280: 276: 269: 265: 262: 246: 243: 240: 232: 219: 215: 211: 207: 203: 200: 196: 193: 189: 188:sawtooth wave 185: 182: 178: 174: 170: 169: 168: 162: 153: 151: 147: 143: 142:Kevin Karplus 139: 135: 133: 129: 125: 121: 117: 116:feedback loop 113: 108: 106: 102: 98: 94: 84: 81: 73: 70:December 2013 63: 59: 53: 52: 46: 41: 32: 31: 19: 961: 897:Sample-based 764: 715: 688: 682: 648: 642: 635:Bibliography 621:. Retrieved 617: 608: 597: 585:|title= 565: 532: 497: 493: 489: 484: 480: 476: 472: 460: 458: 443: 420: 412: 403: 390: 354: 348: 338: 334: 330: 304: 298: 294: 287: 282: 278: 274: 267: 263: 228: 166: 156:How it works 136: 109: 92: 91: 76: 67: 48: 877:Subtractive 445:Alex Strong 423:Alex Strong 350:Z-transform 333:(4.2) = 0.8 261:phase delay 192:square wave 173:white noise 150:portmanteau 124:z-transform 120:comb filter 114:based on a 62:introducing 887:Distortion 623:2019-07-17 519:References 406:media help 199:delay line 105:percussion 45:references 909:Wavetable 524:Citations 449:wavetable 343:harmonics 337:(4) + 0.2 247:π 241:− 181:sine wave 128:wavetable 1028:Category 914:Granular 882:Additive 507:See also 177:wideband 985:Modular 957:Formant 901:Sampler 872:Scanned 705:3680063 665:3680062 559::  396:= 220Hz 208:. The 58:improve 919:Vector 771:  755:  736:  703:  663:  357:Vorbis 293:where 206:filter 101:string 47:, but 847:types 701:JSTOR 679:(PDF) 661:JSTOR 485:Grass 184:chirp 769:ISBN 589:help 425:and 229:The 210:gain 146:VLSI 122:for 899:or 693:doi 653:doi 471:'s 323:FIR 321:or 319:IIR 313:), 190:or 107:. 1030:: 699:. 687:. 681:. 659:. 647:. 616:. 580:: 578:}} 574:{{ 543:^ 277:= 837:e 830:t 823:v 777:. 707:. 695:: 689:7 667:. 655:: 649:7 626:. 591:) 587:( 408:. 394:1 391:F 339:s 335:s 331:s 299:s 295:F 291:0 288:F 286:/ 283:s 279:F 275:D 271:0 268:F 264:D 244:2 194:. 83:) 77:( 72:) 68:( 54:. 20:)

Index

Karplus–Strong algorithm
references
inline citations
improve
introducing
Learn how and when to remove this message
physical modelling synthesis
string
percussion
subtractive synthesis
feedback loop
comb filter
z-transform
wavetable
digital waveguide synthesis
Alexander Strong
Kevin Karplus
VLSI
portmanteau

white noise
wideband
sine wave
chirp
sawtooth wave
square wave
delay line
filter
gain
positive feedback

Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.