Talk:Constant-Q transform

This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Statistics Mid‑importance This article is within the scope of WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.StatisticsWikipedia:WikiProject StatisticsTemplate:WikiProject StatisticsStatistics articles Mid This article has been rated as Mid-importance on the importance scale. Signal Processing This article is within the scope of WikiProject Signal Processing, a collaborative effort to improve the coverage of signal processing on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Signal ProcessingWikipedia:WikiProject Signal ProcessingTemplate:WikiProject Signal ProcessingSignal Processing articles Mathematics Low‑priority This article is within the scope of WikiProject Mathematics, a collaborative effort to improve the coverage of mathematics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.MathematicsWikipedia:WikiProject MathematicsTemplate:WikiProject Mathematicsmathematics articles Low This article has been rated as Low-priority on the project's priority scale.

In my opinion at least, I think the page looks a little rough could do with some better formatting... But right now it's 4am, I need sleep, and I'll be working on my project for a bit :-(

Intelmole 04:04, 15 March 2007 (UTC)[reply]

i am not shure about "j" in exp function, what is it? —Preceding unsigned comment added by 77.184.102.91 (talk) 20:49, 11 November 2007 (UTC)[reply]

Apologies for my electronic engineering nature, j = i = sqrt (-1) Intelmole (talk) 23:28, 28 January 2008 (UTC)[reply]

Can we add a definition for S? — Preceding unsigned comment added by Wavelet transformer (talk • contribs) 07:07, 10 August 2017 (UTC)[reply]

The waveform representation of the piano note, with the caption "Its waveform does not communicate pitch": yes, it does, and quite clearly. But, it's too scrunched up in the picture, and needs to be blown up to a larger scale, where you see the big wigglies that make up the wave - about 4 milliseconds in width (or about 250 Hz), and then the little wigglies on the big wigglies (the higher harmonics). It's there, and clearly in view. You just need to know what to look for, and how to correctly look at it. The picture is misleading because it's shown at the wrong scale. This is generally the case for other waveforms. The lowest frequencies show up quite clearly, once the waveform is shown at the right resolution. The next higher frequencies show up as secondary waveforms superposed on the primary waveform, showing up clearly when you increase the resolution further, and so on. — Preceding unsigned comment added by 2603:6000:AA00:747B:0:0:0:193B (talk) 19:00, 8 July 2022 (UTC)[reply]

@2603:6000:AA00:747B:0:0:0:193B ok good point. Feel free to write it better or delete it. and maybe I should use a shorter duration of a chord, like a tenth of a second, so that all the squiggles of the waveform can be seen... Em3rgent0rdr (talk) 00:42, 9 July 2022 (UTC)[reply]

I came to this page to learn something about the Constant Q transform, at least to see its essential definition.

The section Constant_Q_transform#Calculation_of_the_Transform is already in trouble at these statements.

${\displaystyle Q={\frac {f_{k}}{\delta f_{k}}}}$

• Window length for the k-th bin

${\displaystyle N[k]=\left({\frac {f_{s}}{\delta f_{k}}}\right)=\left({\frac {S}{f_{k}}}\right)Q}$

As S/f_k is the number of samples processed per cycle at frequency f_k,

It is unavoidable that ${\displaystyle S=f_{s}}$. And somewhere ${\displaystyle f_{k}}$ must be defined in terms of ${\displaystyle k}$. I cannot see that it is. Then we can define ${\displaystyle N[k]}$. And a floor function will be needed, I think.

I am certain that the mathematics presented in this section are flawed and I think the reason is some lack of rigor. I'll have to find out the definition from somewhere else and fix the math here. I'm pretty sure I can and have surmised the definition of the Constant Q Transform, but I will check with other sources before I fix the definition here. 70.109.188.75 (talk) 04:56, 3 June 2014 (UTC)[reply]

Thanks for your concern and willingness to help! Have you read the two papers by Judith Brown that are cited in the article? To the best of my knowledge they are the foundation of the transform this article is attempting to be about, and my guess is that there is no clearer statement of the transform anywhere in the literature. I don't doubt that there are problems with the math here, and I haven't systematically compared it to the papers myself yet. But we also don't know for sure that there aren't problems with the math in the original papers. Eflatmajor7th (talk) 22:09, 3 June 2014 (UTC)[reply]

So I cited the JavaScript source code (especially the inside of calcSpectrum2() function) as a reference for variable-Q bandwidth calculation, which @Masem said it is actually WP:NOR to cite the source code where no papers or other reliable sources exist on this reliable sources noticeboard, therefore I was actually wrong about Wikipedia's purpose especially regarding original research. 180.253.8.202 (talk) 04:38, 6 December 2022 (UTC)[reply]