43:
227:, which are sensitive to any small changes in the data. Acoustic fingerprints are more analogous to human fingerprints where small variations that are insignificant to the features the fingerprint uses are tolerated. One can imagine the case of a smeared human fingerprint impression which can accurately be matched to another fingerprint sample in a reference database; acoustic fingerprints work similarly.
261:
techniques will make radical changes to the binary encoding of an audio file, without radically affecting the way it is perceived by the human ear. A robust acoustic fingerprint will allow a recording to be identified after it has gone through such compression, even if the audio quality has been
289:
Any piece of audio can be translated into a spectrogram. Each piece of audio is split into segments over time. In some cases, adjacent segments share a common time boundary, in other cases adjacent segments might overlap. The result is a graph that plots three dimensions of audio: frequency vs
222:
A robust acoustic fingerprint algorithm must take into account the perceptual characteristics of the audio. If two files sound alike to the human ear, their acoustic fingerprints should match, even if their binary representations are quite different. Acoustic fingerprints are not
499:
301:'s algorithm picks out points where there are peaks in the spectrogram which represent higher energy content. Focusing on peaks in the audio greatly reduces the impact that
475:
A Review of
Algorithms for Audio Fingerprinting (P. Cano et al. In International Workshop on Multimedia Signal Processing, US Virgin Islands, December 2002)
190:
identification. Media identification using acoustic fingerprints can be used to monitor the use of specific musical works and performances on
398:
506:
107:
309:, where the key is the frequency. They do not just mark a single point in the spectrogram, rather they mark a pair of points: the
79:
17:
536:
317:. So their database key is not just a single frequency, it is a hash of the frequencies of both points. This leads to fewer
60:
86:
609:
93:
126:
604:
75:
64:
335:
251:
340:
541:
492:
479:
480:
Content-Based
Retrieval of Music and Audio by Jonathan Foote, ISS, National University of Singapore.
386:
Multimedia framework (MPEG-21) -- Part 11: Evaluation Tools for
Persistent Association Technologies
100:
583:
148:
144:
53:
402:
258:
450:
578:
474:
425:
210:
networks. This identification has been used in copyright compliance, licensing, and other
8:
561:
267:
566:
355:
345:
298:
31:
573:
551:
515:
365:
243:
175:
556:
302:
360:
279:
263:
230:
Perceptual characteristics often exploited by audio fingerprints include average
203:
191:
546:
318:
247:
239:
179:
160:
305:
has on audio identification. Shazam builds their fingerprint catalog out as a
598:
231:
224:
187:
531:
350:
211:
207:
183:
156:
152:
330:
283:
306:
266:
monitoring, acoustic fingerprints should also be insensitive to analog
484:
42:
282:. One common technique is creating a time-frequency graph called a
171:
166:
Practical uses of acoustic fingerprinting include identifying
235:
195:
167:
27:
Condensed digital summary generated from an audio signal
278:
Generating a signature from the audio is essential for
199:
30:
For acoustic emissions of ships and submarines, see
67:. Unsourced material may be challenged and removed.
596:
423:
500:
427:An Industrial-Strength Audio Search Algorithm
321:improving the performance of the hash table.
399:"How does Shazam work to recognize a song?"
507:
493:
127:Learn how and when to remove this message
14:
597:
537:Computational auditory scene analysis
514:
488:
396:
159:or quickly locate similar items in a
65:adding citations to reliable sources
36:
397:Surdu, Nicolae (January 20, 2011).
155:, that can be used to identify an
24:
262:reduced significantly. For use in
246:, prominent tones across a set of
143:is a condensed digital summary, a
25:
621:
468:
41:
290:amplitude (intensity) vs time.
52:needs additional citations for
443:
417:
390:
378:
273:
13:
1:
371:
336:Automatic content recognition
217:
384:ISO IEC TR 21000-11 (2004),
341:Digital video fingerprinting
7:
542:Music information retrieval
324:
10:
626:
29:
610:Fingerprinting algorithms
522:
293:
186:library management; and
605:Acoustic fingerprinting
584:3D sound reconstruction
18:Acoustic fingerprinting
76:"Acoustic fingerprint"
579:3D sound localization
433:, Columbia University
424:Li-Chun Wang, Avery,
527:Acoustic fingerprint
141:acoustic fingerprint
61:improve this article
562:Speaker recognition
145:digital fingerprint
567:Speech recognition
451:"How Shazam Works"
356:Perceptual hashing
346:Feature extraction
280:searching by sound
151:generated from an
32:Acoustic signature
592:
591:
574:Sound recognition
552:Speech processing
516:Computer audition
453:. 10 January 2009
366:Sound recognition
259:audio compression
244:spectral flatness
149:deterministically
137:
136:
129:
111:
16:(Redirected from
617:
557:Speech analytics
509:
502:
495:
486:
485:
462:
461:
459:
458:
447:
441:
440:
439:
438:
432:
421:
415:
414:
412:
410:
401:. Archived from
394:
388:
382:
303:background noise
234:rate, estimated
132:
125:
121:
118:
112:
110:
69:
45:
37:
21:
625:
624:
620:
619:
618:
616:
615:
614:
595:
594:
593:
588:
518:
513:
471:
466:
465:
456:
454:
449:
448:
444:
436:
434:
430:
422:
418:
408:
406:
395:
391:
383:
379:
374:
361:Search by sound
327:
319:hash collisions
296:
276:
264:radio broadcast
248:frequency bands
220:
204:streaming media
192:radio broadcast
133:
122:
116:
113:
70:
68:
58:
46:
35:
28:
23:
22:
15:
12:
11:
5:
623:
613:
612:
607:
590:
589:
587:
586:
581:
576:
571:
570:
569:
564:
559:
549:
547:Semantic audio
544:
539:
534:
529:
523:
520:
519:
512:
511:
504:
497:
489:
483:
482:
477:
470:
469:External links
467:
464:
463:
442:
416:
389:
376:
375:
373:
370:
369:
368:
363:
358:
353:
348:
343:
338:
333:
326:
323:
313:plus a second
311:peak intensity
295:
292:
275:
272:
225:hash functions
219:
216:
180:advertisements
161:music database
135:
134:
49:
47:
40:
26:
9:
6:
4:
3:
2:
622:
611:
608:
606:
603:
602:
600:
585:
582:
580:
577:
575:
572:
568:
565:
563:
560:
558:
555:
554:
553:
550:
548:
545:
543:
540:
538:
535:
533:
530:
528:
525:
524:
521:
517:
510:
505:
503:
498:
496:
491:
490:
487:
481:
478:
476:
473:
472:
452:
446:
429:
428:
420:
405:on 2016-10-24
404:
400:
393:
387:
381:
377:
367:
364:
362:
359:
357:
354:
352:
349:
347:
344:
342:
339:
337:
334:
332:
329:
328:
322:
320:
316:
312:
308:
304:
300:
291:
287:
285:
281:
271:
269:
265:
260:
255:
253:
249:
245:
241:
237:
233:
232:zero crossing
228:
226:
215:
213:
209:
205:
201:
197:
193:
189:
185:
181:
177:
173:
169:
164:
162:
158:
154:
150:
146:
142:
131:
128:
120:
109:
106:
102:
99:
95:
92:
88:
85:
81:
78: –
77:
73:
72:Find sources:
66:
62:
56:
55:
50:This article
48:
44:
39:
38:
33:
19:
532:Audio mining
526:
455:. Retrieved
445:
435:, retrieved
426:
419:
407:. Retrieved
403:the original
392:
385:
380:
351:Parsons code
315:anchor point
314:
310:
297:
288:
277:
268:transmission
256:
229:
221:
212:monetization
208:peer-to-peer
184:sound effect
165:
157:audio sample
153:audio signal
140:
138:
123:
114:
104:
97:
90:
83:
71:
59:Please help
54:verification
51:
409:12 February
331:Chromaprint
284:spectrogram
274:Spectrogram
270:artifacts.
599:Categories
457:2018-04-02
437:2018-04-02
372:References
307:hash table
238:, average
218:Attributes
188:video file
87:newspapers
252:bandwidth
214:schemes.
117:June 2011
325:See also
240:spectrum
172:melodies
196:records
101:scholar
299:Shazam
294:Shazam
250:, and
206:, and
103:
96:
89:
82:
74:
431:(PDF)
257:Most
236:tempo
178:, or
176:tunes
168:songs
108:JSTOR
94:books
411:2018
80:news
200:CDs
139:An
63:by
601::
286:.
254:.
242:,
202:,
198:,
194:,
182:;
174:,
170:,
163:.
147:,
508:e
501:t
494:v
460:.
413:.
130:)
124:(
119:)
115:(
105:·
98:·
91:·
84:·
57:.
34:.
20:)
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.