[Rwp] Accessible spectrum analyser
Fiore Martin
mister.bloom at gmail.com
Thu Mar 3 05:00:06 EST 2016
Hi everybody,
Chris:
it is exactly as you said: the higher the frequency the further to the
right it is, and lower the frequency it goes to the left.
Where did you spot the mistake ? In the webpage of the analyser I am saying
that :
the sonification is *panned from left to right* and the panning represents *the
position of the peaking frequency* in the whole spectrum, ranging from 20
Hz on the very left, to 20050 Hz on the very right. For example, if the
selected spectrum peaks at 50 Hz then you'll hear the beep towards the
left, whereas if the peak is at 20 kHz the beep will occur towards the left.
Maybe not the clearest way to put it but I think here we were on the same
page.
Also : it's not possible to have the sonification going through a different
sound card than the sound being analysed. However you might find useful to
play with the separate wet and dry parameters to go some way into that.
Hopes this is helpful.
Thank you for your feedback.
This project is now terminated so I am officially no longer working on this
any more.
I know that for those of you who are not programmers it doesn't make much
difference but we released the source code as well.
It's linked from the web page of the plug-in
all the best
Fiore
2016-03-03 8:08 GMT+00:00 Chris Belle <cb1963 at sbcglobal.net>:
> Well, the last point could be done by a time selection in your daw,
> and the table set to generate till the daw stops or something like that.
> Like a timed run of a specific portion.
> leave it to your brilliant mind to analyze and break all this down
> I like that table thing,
> and being able to change the resolution of what is analyzed would be great.
> I'm going to persevere with the current implementation and see what I can
> do with it.
> This is the first tool of it's kind I believe,
> we've had audio meters in hardware before, but it's nice now that the
> software ones are finally coming around to reality instead of just concept.
> I'm not a programmer, but in theory it shouldn't be hard to generate
> some kind of values which can be tied to text out-put,
> we'll never have it s good as being able to look at a graph, but it
> should be possible to link up something useful can you imagine as a
> programming guy how this might be done?
> YOu and this fellow would make a formidable programming team I bet.
> 'grin'.
>
>
>
>
> On 3/2/2016 2:20 PM, Snowman wrote:
>
>> Well, here are my thoughts about this, for what it's worth.
>> I find myself less able to make use of a real-time, dynamic audible
>> output like this, especially while I'm also hearing the audio itself. But,
>> also in general.
>> To be honest, I wish this tool could generate a spectrum report, which
>> would be a table of numbers describing a curve, showing how much peak, and
>> average energy is in each frequency band, so you could study it, and learn
>> where the peaks really reside.
>> A constantly changing tone output has time delays associated with it, as
>> you are aanalyzing the selection in real time. So, yeah, it takes a bit of
>> concentration, especially for a partially tone deaf person, like me. I
>> know, Not a musician any more. But, I still do audio production.
>> If the tool would let me specify a bin size, maybe in hertz, or in
>> octives, and the start and end frequencies of the spectrum I want to check
>> out, much as it does now, except for the addition of the bin size. And
>> then, produce a table of amplitudes, which generally show how much energy
>> is in each bin, then I could understand what is actually going on.
>> As it is, you have to violate the threshold before you get any indication
>> that some coloration still exists.
>> So, for example, I would specify start frequency, then width, or end
>> frequency, and then bin size.
>> So, Maybe I watch the area between 1k and 11khz, for a 10K total span,
>> and I specify 100 hz bin size. so we divide the 10k span by the 0.1khz bin
>> size, and then we generate 100 values. The first value represents the
>> amount of energy in the span from 1k to 1.1khz. By inspecting that table,
>> manually, or witt computerized aid, I can find out which bin has the most
>> energy, even if it did not violate any threshold. From that, I can rather
>> precisely tell where I am sharp.
>>
>> The last thin is to somehow identify the portion of a track, or file,
>> that should be analyzed.
>>
>> Anyway, that is what I hope for. More of a static analysis tool, instead
>> of a real-time dynamic tool.
>>
>> I don't claim that the current implementation doesn't have value. I think
>> it does. I just wish for this additional static data output.
>>
>> ----- Original Message ----- From: "Chris Belle" <cb1963 at sbcglobal.net>
>> To: "Reapers Without Peepers" <rwp at bluegrasspals.com>
>> Sent: Wednesday, March 02, 2016 7:28 AM
>> Subject: Re: [Rwp] Accessible spectrum analyser
>>
>>
>> I think there's some slight errors in the documentation,
>>> you said left when you meant right,
>>> or maybe I don't understand correctly, but
>>> the way I grasp it is that the tone goes up in pitch as the volume of
>>> the peak goes up, and the position of the sound in the
>>> pan field changes with frequency, the higher the frequency the further
>>> to the right it is, and lower the frequency it goes to the left.
>>> Is that right?
>>>
>>> And you can select the starting point, and the band width of the
>>> frequency range that is being tested.
>>> I'm curious,
>>> can the beeps be separated from the sound of the audio your testing?
>>> Right now the audio jumps around in the pan field as well, and that is
>>> rather disconcerting if you are trying to listen to it,
>>> and find a hot frequency spot as well as hearing the beeps.
>>> So if it's possible to unlink the audio and the beeps it might be easier
>>> to work with.
>>> Just a thought, it may not bother others.
>>> Thanks for working so hard on this.
>>>
>>>
>>> On 3/2/2016 6:40 AM, Fiore Martin wrote:
>>>
>>>>
>>>>
>>>> 2016-03-02 10:49 GMT+00:00 Chris Belle <cb1963 at sbcglobal.net <mailto:
>>>> cb1963 at sbcglobal.net>>:
>>>>
>>>> Hey this is wonderful.
>>>>
>>>>
>>>> Thanks!
>>>>
>>>> You're the guy who also developed the accessible peak meter thing
>>>> too right?
>>>>
>>>>
>>>> That's right. In fact the spectrum analyser came from the idea of using
>>>> the accessible peak meter's very sonification into the frequency domain.
>>>>
>>>>
>>>> If you guys haven't gotten that tool, it's really useful as well.
>>>>
>>>>
>>>> On 3/1/2016 9:44 AM, Fiore Martin wrote:
>>>>
>>>> Hi all,
>>>> just to let you know the Accessible Spectrum Analyser VST/AU
>>>> plug-in is now available online for free at
>>>> http://depic.eecs.qmul.ac.uk/apm/spectrum.html
>>>>
>>>> Accessible Spectrum Analyser makes spectrograms accessible
>>>> using non speech sound to supports core activities in audio
>>>> production.
>>>> The plug-in was developed as part of our research on
>>>> accessibility and cross modal collaboration at Centre for
>>>> Digital Music, Queen Mary University of London.
>>>>
>>>> More info on the web page linked above
>>>>
>>>> thank you
>>>> all the best
>>>> Fiore Martin
>>>>
>>>>
>>>> _______________________________________________
>>>> RWP mailing list
>>>> RWP at bluegrasspals.com <mailto:RWP at bluegrasspals.com>
>>>> http://bluegrasspals.com/mailman/listinfo/rwp
>>>>
>>>>
>>>> _______________________________________________
>>>> RWP mailing list
>>>> RWP at bluegrasspals.com <mailto:RWP at bluegrasspals.com>
>>>> http://bluegrasspals.com/mailman/listinfo/rwp
>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> RWP mailing list
>>>> RWP at bluegrasspals.com
>>>> http://bluegrasspals.com/mailman/listinfo/rwp
>>>>
>>>
>>> _______________________________________________
>>> RWP mailing list
>>> RWP at bluegrasspals.com
>>> http://bluegrasspals.com/mailman/listinfo/rwp
>>>
>>>
>>
>> _______________________________________________
>> RWP mailing list
>> RWP at bluegrasspals.com
>> http://bluegrasspals.com/mailman/listinfo/rwp
>>
>>
> _______________________________________________
> RWP mailing list
> RWP at bluegrasspals.com
> http://bluegrasspals.com/mailman/listinfo/rwp
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://bluegrasspals.com/pipermail/rwp/attachments/20160303/c8462083/attachment-0001.html>
More information about the RWP
mailing list