Geek Answers Handbook

Audio Analysis: Frequency and Step

I am designing a simple tuner, so my goal displays the name of the note (A, B, F #) and the distance in cents between the theoretical sound and the actual input.

I am completely new to audio and signal processing, so I did some research and I found a thing called Fast Fourier Transform that will analyze bytes and give me the frequency. In addition, I found a couple of Java libraries such as general math and JTransforms , so I myself will not write hard code.

I believed that everything, since each frequency of the range can be directly compared with a note in equal temperament, but then I found this new (me) word called pitch : he said that it is tightly connected with the frequency, but not quite the same and much more difficult to obtain, and belongs to this psychoacoustic field.

So my question is: can someone clearly describe the differences between pitch and frequency and maybe tell me which tuners they are dealing with?

+5
source share
7 answers

Frequency - , . . , , - x, .

, , ( ). , , , . () , , , .

, , . . , , ,

  • 440 1
  • 1320 1/2
  • 2200 1/3

.. , , ,

  • 440 1
  • 880 1/2
  • 1320 1/4
  • 1760 1/8

.. ( , ). , -, , , , . , ; , " 440 ".

pitch, . - , , . , . , , .

, , . , , , . , 440 , , 440 , , , .

+14

- . , , , , .

- . , , "" , , . , , , . " ", , " ", , . , , , , , - .

( DFT/FFT) - , . , , , Spectrum. - , .

, , , , : fooobar.com/questions/198346/... , , .

+5

, , .

, . ( ) , "".

, , , , .

, . (, , 1-), . , . , .

, () , , , .

. FFT , , , , . .

Pitch FFT , , , , , .

: , .

+2

- , . A 440 , , 441, 442,... A ( 440 449, , 1 ). A 880, 882... . ( ), , . . : http://members.efn.org/~qehn/global/building/cents.htm

+1

, "", , A4, , "", 4 440 ( .)

. 440 , - 880, 1320, 1760 ..

, .

:

, , ( ). , .

FFT time/sound_pressure /_magnitude.

FFT " ", "" . , "" , "Un-Fold w".

FFT ( ). "" . .

, "FFT Y-Axis Magnitude" "Sqrt (R ^ 2 + ^ 2)" .

, "dB Norm Sqrt (R ^ 2 + ^ 2)" .

"", , .

"Inv-FFT", /, , .

, Inv-FFT . time/sound_pressure /_magnitude.

"FFT", .

, . , . , unzoomed.

:

  • ( ) FFT.
  • .
  • , , A4 440 .
  • .

, :

  • .
  • ( ).
  • ( ).
+1

"" "".

"" , , , . . , , 261,6 , , 523 , 785 , 1046 - 4- .. , 261,6 (: 2 261,6 = 523, 3 261,6 = 785, 4 261,6 = 1046).

3 MP3. , , . , , . ( DFT)

"", , .

https://en.wikipedia.org/wiki/Transcription_(music)#Pitch_detection

+1

, .

Hertz, , . , 440 , 440 .

, - 1/100 , - 1/12 . , . , " " , .

, - , , . , , , -2 -2 , . "A C" - 440 .

. , , 1000 . -2 1000 9.9657842847. -2 440 8,7813597135. 1,1844245711; , , 1000 , 1,1844245711 "A C" . 12, - 14.21309485 . 14 "A C" - "B" 2 C. , , 21,309485 "B".

, 2 , ( B - 2 A), 1 ( C - 1 B). ( 2 G H, A). ; .

, , , , , . " ", ( "" "" ). , "A C" , , 440 , 880 , 1320 .. - , , 440 . , , , , "pitch", ( "A C" ), , .

, "pitch", ; .. , , , . , , , , . ( ), , , ,

  • , , , , (, , ).

, , , 262 , 440 , 524 , 786 , 880 , 1048 1320 , . , 262 440 . , " " ( C ). , , , . , ; .

To such an extent, the step perceived by the musician is a psychological effect, which complicates the modeling in the electronic tuner. You may need to do some experimentation to figure out exactly when the higher step should be considered as a separate note, and when it should be considered harmonious. In addition, many musicians will be able to hear sounds that do not analyze Fourier analysis (summing tones and difference tones) - their hearing really plays tricks.

-1
source

More articles:


All Articles