IPB

Welcome Guest ( Log In | Register )

Human hearing beats FFT
ncdrawl
post Feb 9 2013, 15:20
Post #1





Group: Members
Posts: 37
Joined: 5-October 08
Member No.: 59436



http://phys.org/news/2013-02-human-fourier...-principle.html
Go to the top of the page
+Quote Post
 
Start new topic
Replies
Yaakov Gringeler
post Apr 11 2013, 19:33
Post #2





Group: Members
Posts: 7
Joined: 1-April 13
Member No.: 107483



The paper described the mathematical basis of EST, which uses the ideal case of perfect increasing/decreasing sinusoids.

For realistic data, EST uses different processes, that expect noise.

For audio, the EST process is as follows.
1. Find linear prediction coefficients, preferably using the covariance method and not the auto-correlation method.
2. Create the linear prediction polynomial.
3. Find the roots of the linear prediction polynomial to establish the basis set of an exponential sum function, as described in the paper.
4. Use the samples and the basis set to find the coefficients of the function.

The key point is that linear prediction coefficients and an exponential sum function, are equivalent, with the exponential sum function having the distinct advantage of being an analytic function with a useful structure. The mathematical basis proves this equivalence.

Due to the equivalence, an exponential sum function models an audio signal with the same quality as linear prediction.

You may note that the best lossless audio compressors, like OptimFROG, use linear prediction. This is a strong indication of the power of linear prediction to model audio.

Since EST generates an analytic function, it is suitable for lossy audio compression, as well as other audio applications.

Once EST generated an exponential sum function, you can do the following:
Identify noise elements, using frequency and/or amplitude, and remove them.
Identify inaudible elements, and remove them.
Quantize the coefficients.
Resample the audio signal, both sample rate and sample depth.
And various other things.

Unlike Fourier related methods, which use a predefined basis, EST uses a basis derived from the data.

In short, EST for audio combines the flexibility and usefulness of an analytic function with the modeling power of linear prediction.
Go to the top of the page
+Quote Post

Posts in this topic
- ncdrawl   Human hearing beats FFT   Feb 9 2013, 15:20
- - probedb   QUOTE (ncdrawl @ Feb 9 2013, 14:20) http:...   Feb 9 2013, 15:51
|- - ncdrawl   QUOTE (probedb @ Feb 9 2013, 09:51) QUOTE...   Feb 9 2013, 16:02
|- - probedb   QUOTE (ncdrawl @ Feb 9 2013, 15:02) it is...   Feb 11 2013, 09:26
- - lvqcl   From the comments: QUOTE The Fourier uncertainty ...   Feb 9 2013, 16:10
- - greynol   That title? What about the title of this discussi...   Feb 9 2013, 18:00
- - Alexey Lukin   Love the title   Feb 9 2013, 20:13
- - greynol   ...with a blunt stick.   Feb 9 2013, 20:17
- - Woodinville   Ad the Gabor limit does not address known frequenc...   Feb 10 2013, 04:41
|- - Kees de Visser   QUOTE (Woodinville @ Feb 10 2013, 04:41) ...   Feb 10 2013, 08:31
|- - Woodinville   QUOTE (Kees de Visser @ Feb 9 2013, 23:31...   Feb 10 2013, 12:51
|- - Porcus   QUOTE (Kees de Visser @ Feb 10 2013, 08:3...   Feb 10 2013, 17:15
- - 2Bdecided   Is it worth spending $25 to read this in full...   Feb 11 2013, 10:25
|- - Kees de Visser   I just stumbled upon this blog. Are we heading tow...   Feb 11 2013, 10:56
||- - Porcus   QUOTE More pointedly: until scientists devise and ...   Feb 11 2013, 11:46
||- - 2Bdecided   QUOTE (Porcus @ Feb 11 2013, 10:46) QUOTE...   Feb 11 2013, 12:21
|- - ojdo   QUOTE (2Bdecided @ Feb 11 2013, 17:25) Is...   Feb 11 2013, 17:13
- - db1989   Given that the writer misspelled mathematical and ...   Feb 11 2013, 16:39
|- - Ethan Winer   QUOTE (db1989 @ Feb 11 2013, 10:39) Given...   Feb 11 2013, 23:05
- - greynol   Would someone mind correlating the article with th...   Feb 11 2013, 16:48
- - benski   Off the top of my head. I might be wrong. But pa...   Feb 11 2013, 18:32
- - 2Bdecided   By taking an FFT, you are performing an (approxima...   Feb 11 2013, 19:02
- - krabapple   Here's what the audiophilosphere reliably take...   Feb 12 2013, 07:18
- - knutinh   Is not some wavelet/filterbank transform more rele...   Feb 12 2013, 09:59
|- - Garf   QUOTE (knutinh @ Feb 12 2013, 09:59) Is n...   Feb 19 2013, 17:27
|- - jmvalin   QUOTE (Garf @ Feb 19 2013, 11:27) Yes, th...   Feb 19 2013, 21:07
- - scuttle   I thought it would be fun and informative to poke ...   Feb 17 2013, 17:33
- - Willakan   Good God, I have been reading/getting PMed on othe...   Feb 18 2013, 20:03
|- - Porcus   QUOTE (Willakan @ Feb 18 2013, 20:03) or ...   Feb 18 2013, 20:22
|- - Woodinville   QUOTE (Willakan @ Feb 18 2013, 11:03) Goo...   Feb 19 2013, 00:44
- - 2Bdecided   It's amazing what people conclude, given that ...   Feb 19 2013, 10:53
- - lvqcl   I didn't find any mention of FFT in this artic...   Feb 19 2013, 15:51
|- - greynol   QUOTE (lvqcl @ Feb 19 2013, 06:51) I didn...   Feb 19 2013, 16:56
- - Canar   The first thing that comes to my mind is: Now how ...   Feb 19 2013, 16:46
|- - Garf   QUOTE (Canar @ Feb 19 2013, 16:46) The fi...   Feb 19 2013, 17:32
- - Paulhoff   For those in the know.......... The Princess an...   Feb 20 2013, 19:51
- - Woodinville   It's still a confused headline. Recognzing one...   Feb 21 2013, 04:55
|- - Garf   QUOTE (Woodinville @ Feb 21 2013, 04:55) ...   Feb 21 2013, 07:54
|- - Woodinville   QUOTE (Garf @ Feb 20 2013, 22:54) QUOTE (...   Feb 22 2013, 04:22
- - dhromed   Yeah, we don't want to turn HA into an extensi...   Feb 21 2013, 10:02
- - 2Bdecided   Never mind the title, I still don't find a sat...   Feb 25 2013, 13:14
|- - knutinh   QUOTE (2Bdecided @ Feb 25 2013, 13:14) Ne...   Feb 25 2013, 15:32
||- - 2Bdecided   QUOTE (knutinh @ Feb 25 2013, 14:32) I gu...   Feb 25 2013, 17:24
|- - jmvalin   QUOTE (2Bdecided @ Feb 25 2013, 07:14) Ne...   Feb 26 2013, 00:10
- - Woodinville   There are a number of issues confused in this thre...   Feb 26 2013, 04:30
|- - Alexey Lukin   QUOTE (Woodinville @ Feb 25 2013, 23:30) ...   Mar 14 2013, 17:39
- - 2Bdecided   Thank you JJ.   Feb 26 2013, 10:57
- - krabapple   http://arstechnica.com/science/2013/02/hum...3s-so...   Feb 27 2013, 01:51
- - greynol   Well the quality of the comments look pretty encou...   Feb 27 2013, 02:44
|- - Woodinville   QUOTE (greynol @ Feb 26 2013, 17:44) Well...   Feb 27 2013, 05:55
- - 2Bdecided   Interesting paper, thank you.   Mar 15 2013, 12:15
- - Yaakov Gringeler   EST is a new transform that can explain the result...   Apr 2 2013, 01:03
|- - Woodinville   QUOTE (Yaakov Gringeler @ Apr 1 2013, 17...   Apr 2 2013, 23:47
|- - Yaakov Gringeler   EST stands for Exponential Sum Transform and it us...   Apr 3 2013, 01:42
|- - Porcus   I think I could very well use a formula or two ......   Apr 3 2013, 20:34
- - Alexey Lukin   Yaakov, also check out the Reassigned spectrogram ...   Apr 2 2013, 23:59
- - Canar   I think a lot of us here would be interested in a ...   Apr 3 2013, 05:27
|- - Yaakov Gringeler   QUOTE (Canar @ Apr 3 2013, 05:27) I think...   Apr 3 2013, 18:14
- - db1989   If I may guess, I think he means that this site ha...   Apr 3 2013, 18:31
- - Yaakov Gringeler   The following link: http://www.mediafire.com/view...   Apr 3 2013, 22:10
- - Woodinville   Hm. Define "noiseless". Most instrumen...   Apr 11 2013, 11:09
- - Yaakov Gringeler   The paper described the mathematical basis of EST,...   Apr 11 2013, 19:33
|- - Woodinville   QUOTE (Yaakov Gringeler @ Apr 11 2013, 11...   Apr 11 2013, 20:36
|- - Yaakov Gringeler   QUOTE (Woodinville @ Apr 11 2013, 20:36) ...   Apr 11 2013, 21:32
|- - Woodinville   QUOTE (Yaakov Gringeler @ Apr 11 2013, 13...   Jun 4 2013, 01:51
- - Specy   Over 10 years ago, for my master thesis, I wrote a...   Aug 17 2013, 11:52
- - Yaakov Gringeler   Several months ago, in posts in this topic, I prov...   Nov 4 2013, 20:15


Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 21st October 2014 - 13:03