Creating a 2D Corpus Explorer
Here is another video tutorial (as a series of 4 smaller videos) for you FluCoMites to chew on. Code along with me to create a corpus slicing, analysis and exploration patch in Max.
I’d be grateful for any feedback, positive or negative on the tone and content as well as any indications if you got stuck or I wasn’t clear about something.
VIDEO
Enjoy!
4 Likes
mrhay
December 17, 2021, 9:42am
2
For me this was a solid 4 part introduction and I’m very thankful for it and I would advise that you do more but I’m well aware of what a huge task these things can be to produce.
A couple of things, I did comment on the YouTube videos, but comments there are often like the Wild West…
The linked .maxpat files lead to a 404 error apart from the 4th version.
When I downloaded and pointed Max at the latest version and loaded this 4th version I got some missing objects. It’s probably me forgetting to do something silly and easy, it usually is but I thought I’d ask here and say hi anyway!
Hey @mrhay and welcome
Thanks for engaging with the series, I’ll address your feedback inline.
This should be fixed now as I’ve updated the links. Can you let me know if it is not?
It looks like you’re on a Mac. can you confirm for me that the package is in:
/Users/yourusername/Documents/Max 8/Packages
for me?
This problem usually arises when fluid.libmanipulation.mxo
doesn’t load up properly because its sensitive to where it lives.
Welcome again!
1 Like
this is true - and also that it needs an init file to map all the objects to it - it is a library in a box!
1 Like
mrhay
December 17, 2021, 9:15pm
5
Ok right cool and thanks for the response. I wasn’t aware that the package was a little fussy. I plonked it there and it seems fine right now!
And yes I can confirm that the links now lead straight to a download prompt. So nice one for sorting that out. I’ll get digging and explore.
P.S. If Dr. Green is still with you lot, please say hi from me.
Stay well out there.
1 Like
Hey, Kev Hope all’s hanging well in the antipodes xx
2 Likes
I think I’ve found a small bug in corpus-explorer.maxpat :
actually analysis is limited to 256 segments :
In « p analysis » fluid.buf2list 2048 @source slicepoints outputs a list with up to 2048 elements, but below there’s a zl.stream 2 which is missing a @zlmaxsize 2048 argument… so it clips the analysis to the first 256 elements.
2 Likes
I should fix that - I find a few of these but corrected locally… thanks for bringing them back to my attention!
I isolated 2 more small issues in corpus-explorer.maxpat
As I based my own project on this patch & tutorial (and I guess many new FluCoMa users do the same), I think it’s a good idea to fix it
Number of slices != number of points
With default Nicol-LoopE-M.wav : 24 slices are found
But there are only 23 points in the dataset analysis.
The culprit is zl.stream (him again!) (in analysis)
If you send a list of N elements to zl.stream 2, it will output N-1 pairs of values.
That means that the last slice won’t be analysed.
To fix it, one value should be added and the end of the slicepoints list : the easy way is to add a value (in samples) corresponding to the end of the sound file.
OK, now we have all slices analysed.
But that leads to another issue in the playback patcher :
The last slice cannot be played , because the length of the segment is calculated by getting the position of the next slice in slicepoints buffer. But for the last slice, there is no next slicepoint…
so in that case, we should also use the end of the buffer as pseudo-next slice position.
To do that in the playback patcher, we need to know the number of slices, and the length of the buffer.
Here is a modified version of corpus-explorer.maxpat, fixing these 2 issues.
For the playback / last slice issue, I added a send & receive « slicesCount » in order to get the number of slices in the playback patcher.
Another way to fix both issues in one shot would be to manually add a “fake” slicepoint, that would represent the end of the buffer.
Analysis blocks and points count would then be correct, and no issue in playback.
Is there an elegant way to add a sample in buffer~ slicepoints ? (buffer size must be modified without deleting buffer content…)
Here is the patch :
----------begin_max5_patcher----------
4629.3oc6c08aiiaD+4j+JTMtGZulMmHEo9nObHGZ6ATf16kBz9Puh.Ya4Ds
qrjqjb1r6gK+sWRNT1RNRVT1hJ710YwFGpOrF9SCGNyvYF9KWe0r4YOGULy5
OY8ert5pe45qtRbH9AtR19pYqCedQRXg3xlsLOayp3jnY2.mLc65rskIQkhS
ikGMdo3hyl+92QqtR3xJ+zlH34Ma1Mr+a8ekmdSX4hGiSe397nEkvUfvd2Ze
iElx+MxEw+fxNzt6g8viSqd1H9w90qul+qaTr2rHa85nzxJRrL5Ywid1eIO7
AqPqUYIKixsVjkVFFmxHNqmBShWZ8C+se7Gsxxs92+v+xhCFEVOFkuCRRhSi
VjsMU7U4HO37symmH54nVQN6WibjYcCM3.GAz3KfFhnAEMlPSZzGYDQEI7TX
dZ3ZA8OaUx13k2x.kEgkgaWFmIffWggseYV2sLpnLNMrLNK0pfASKamUB8Z.
InSVokwK3eeg4eREFJDQ.aXb.7g9fsC.i8vPbpx8Zb2CfTouZK5qAjIqqtw5
+sMh8hPUADXmSWBgi6sAhebk+3yFFD3x6qj9d4V+aMJW1OkczqlweK8TTdAi
qp1Ue0rvMapc3qpcKbz48YhuH+a1cn3T3PzcGJO5o3p62d2QCyY8sRVGaaNL
J6Y2pw+7ulLlbnzsw6dUIdOIIIwaD9nyhMgKfal+hq5z6QKZ.vQHvGWhs.l7
s2iPr23Ojjs3CQKqSbyx1DwD9sIOpfIqTLvswoWFsJbaR48qXhIKh+LHjig6
sc9URJr0StSByOjGGlrqC7Pd7xrTNQz3MA+vUONF6NLIAsdmQbEogaZ4lYLd
LXoiSVv5jaKlGlyeQIkZiqNYYVVRySs69RhVUJO8l3zzCPwxrMcex73Gd7H2
67L1IWeruawYJteaJb16Y7Dk2WD9TSztLLIQNrs4W+ygowqCKiJigWAX6cmL
JMj0QerXQdVRRi9KblmZ4LKY73Kh9X7xxGEOn5LCrKOdSESzrcukWF+.ahgl
GqL7ghlGon7S.nW6PamKGCeeYz5MIrdQyKfM5Htnr3wrOVHuvJFs5.vd0fpO
lttLvFG+.YgfnqcCX6Z58Ck8gpeGsHeiJFi5.hxcrq98NV7tjpAxEjJ2TGLD
O+zkQOWSRhTvuTnxIB.MlL3fIDxiXivVZ8HiGOgym2MLgZGlBn0uktmGr64B
gI78.gD1nCmZniID0O1rJOacwmVOOKY3fB47.Ef2RhIdlCjjy5JQVeHMJLmK
PnSXA2Nr3GbbX4F0gFrPsZeZ+PCdDgF173EgOD0J1TgJMTwRcVFemwfkA4Hz
iv2eZwkivxLe6pUQ4u.nh0cEgq2TXgGNqC5nvypjrvRN+y7vzG5WlCn.pDrP
DayQlivVjD1bhXFv0zfrSjux9n.G75QUtKa3Cu.CCv9vxx7nnWrRxx9v1MCl
4xyarjKg7ExkPHmoEgDeqCl2.iFCYNDaU0.x9sRCnSTEPLdlRJunKk.aSD8d
Sn39xpCEiEjJ+7siMEYayWT8VtRINqlTcMIO0tPNhvuPkd2LXpfnHUfzHQv0
QQIhnELac7xMYwokE6romxDB.NGE43sqUiaJKeI3hC6239CQw9CQ3ihZMZu6
fzS2wQwtCWlt9XRrUjJpHWsPEUcw9oh.MREXjpTgcmTg7fU9MaF24HKuGbDz
8gkk4wyYFdTbni.GjCKdHIadXxAdYnM+Yb8dhar755OOa410arRyxWGxTtKZ
o0xvxPqxLqMIYkkQ4+7L084r8o4z417GqCQ36YD3V1KNj8hCYu3P1KNjU2Nj
sWGLwEVNXi3bsOei3PDL3Q.g97tAli2kNNjzg8czyz9Nj..7D1x5DXLtTBr4
mOCZQT4K0lUcv7LjQvgj.HgPfV9.Cjoa3OxeLXL.mb7kng+nfYpInHvYxV8G
sX3uqpFxfzoIDTUMgfnSpP7kiTfJbs0HUvGZpDVPQeEaN0p3RKvouCvxIj2H
Z4jMd+biWLb5hgSWLb5hgSFZjrz2pXHiTUJw3ijkSTa2ySYWY.pX9qx0wrej
OgoJFJ0EDRGCLDLn121vrj7LW8Xj64aDobHHhbVqdrVLQf28TasA0oVwpt.k
UTqdnBUIB5WwplWIkQlEAKX+c5CCQKcxIpkNjPJUq15kPL+PEyc7DRWnNWzL
+hl4WzLeT0L+XpdsION8DTZ09L03BjGVoUO1T8dedzxshDDa3K3yHn1ExEbV
gze1SbT60iB6k4goEqXSmZElFl7IFWrJnUWqEz4wNAF93JS2SeCiaZ65vMuX
cGi.VxDfkxme8TBxXmymixwtNCk+DGYwZxAAPmBSweU5f.kBob46cWi2IASx
5phrGggRUYFjsKORCm7UVUYgy6jIeNNXIveLjOijBn8cLLAzRj4yQuLXdo.2
wRprjSJXhmkmaUOy1sgxQPZJTNmoyZISGWPo6CEgzCCCmDfWTaWmFwr0Xd11
7h9bKLjKdPxdfn1FjOoBTcQRCzo6fBLAOiopiwzYfHSL.ZP03Buh0QONHT4v
ov1w.7S4WutorxfJ08JY2UBiF5GdDOSBJG5hu3XxW4XRHCmgxghui3COOuKN
l7hiIu3XRsEq0EVEILvn3OKpGVidtQFHRPLTUccvbxw5OmbaRTp0ceNgcG7A
pLlJh+vswcDhcXIFA07JjC0XrZK54M4V+9uYE5a+lU3+v2wd8Ye6vcqh64ZS
Kplk+HzDCOLMbJOAGs1L1pGrgaJ4so.wDlvB4gIU+9Mej06yhSm7PKwCPAaP
IKrqwLFJNcU1KMJids.JAmTlarqbWvqTD7O2c.U9C9+4KIkpi+jKRBwaZEg2
r.P1DaQVKdLjoOQh0yrAC7R+31nhgGl+j9b7sKjqyHbcefXa.C0pJqJqRBGd
c3A4LhESEJxuFF4RMrRCBGnXXTYT5KV2AVJasd0hEbcoOnhWdZXI57UCvqNW
FVkh2yjMNyoZfVAajlm0+jalECwVL7Aa8kRMNfceDWaScr1N1lSX8PFwAbj.
AeBoJP6sMuQbBTp43MVCwHtSGD8GqAZRjCSm3kwV8AZ+j0JtZiCePlWOCxHH
Onj838lLH6DUolpeMpANCGP3iAoRccYOib7LLvo4g0TTx43YfSyygn8xbD5c
ac2pnPtqKKjxfXDHeXVgEZvfIdzD+.PniSvuIBNj9pCpvfED1w3CNjwOH8bN
SObfAeZ3BVsZPAoWo07gCFNyFIwvTxtYlMDgLh8Lfn7twDRGXxwkZH34jBOT
znbIB4CUnBEJMrzIZDT3RnXgY8MndrjpqpQn2X3HH4Xo.joFvqGrziCXBH5X
MCjqLLgvdFy.Ll.G1+VjDEly+qtAGZWtzPAoO2r+SwSp9ATR7DLWGjz2HCRG
ne26rNE+yi8UPAw9wDXxKGPoYnr.ZHygYkLbS06YXVE+R+C0n25Q45BYqJtL
U55vL+NuTXfISV8Iv0blE2bYtYws0hVEdYlQQVh8PA6tOMb4Lqf2UCj.bwmZ
PJClX84DoDyA6dz9WlzpuaEWoGXYuntlzBJWTlGEt1Be9qpLZzlkGCKdAxkZ
XpBwrUGyW3JABs2fcdvHHK2rCdrWONXUrLYpgZxU6Bil3Pn9DybA74ayEWoF
1OXaG3m.2Iszdok.DV03CFgLgxZkVSZcAU3pJUfdiCSYcFpzNJGu18VZuwrA
MTtgT1xsgjpF5hzUM7lcv8Q5.wR8w6obQCcQ4XUAcO0nbWTMJWzPWE1ZuQic
QRrTTMJG1RH0CkqZE4VmwfdfIrkBnb0zC4n0.x2cHUNDzwXkbjEZTOgUB3.L
DkkrVu4kdEsViHgWPjATXD6FCQ9PbWZKzLtBQso2pWh2QU423iR7Taf3gRpj
SscRTngTzNBx1McJaG5V3QQDIu.5dzt0jMWKzqPJwwaqcN9258CEgSBUiJvZ
UDJlNrrZ5nyFi4dlCxKGfgRzPqjtRLTs.1GP7AR2JJSNZJP87VS4tJyP5RDU
6RDZ0N.cUqobmkoMB8n1nbDYYxtPfSMFLdCswf4LVjdEC0zQ5lgBZJqtJ5HE
2swhNPF.cnrVeX8RGCs.kOgx.Um1ncSaZRXlptdpEcLZ2eA0mrj14jkeom2v
aRB+z7vEeP87FF0YdCWD+PZXx9LGt8zTs8pEpXQynf+Dbur4McXFE6AU8LHR
58gv8kFPujPwWRn3KITr917lTKghQpjPwmvxmCUnaYPM4S5e46rM+7.D4a.4
AHh.FLiBfvexbBKyhnjwNgsTN9dP.Klr.aRolStjvPEFUZwELBiHs99u+6s1
VDYAYSfkH7DBk4+uUzw3KsUYmit0EU1oF5HqJyFQ1b84ja4o.PyP.ZpR.IXx
aJ3PSyoN3yz3goZvRKQjgM7fP3LCbW.THffaCZukmG2SIict9M.1DGXkNc7m
1PGbziyzfQHJScHfSA8MrfL8jJ4Aix3EGeSaeY7jvBxXfEXYra4GXdUPEquy
hWCUr9Vdp5ZO4EQEoZwxnlDgMmfAbSTzGd47B6OmwH90k1M.J2hrUHwQb9sE
Kj6nLDy7XfNqBrhITeUfTUZGmmiwn4C2Mm8hrmTdxJc5opIbiT0PHhF7HeQj
vrU1.X9IL6Q3P9iVnwtROojqHjFcKybMSJYihY+6H4sliJkfntRPT4GCXHSU
15aNhTN6Y6GiI6kNOWpt3TOW+oEi+3QwbAnve7a2H5W00m2qu.nj.aHNPRbR
.yJ8zW.NnbzSQz3xnqZjLi5MFrg3dmXWU5AqZnIBW03lknFcikU9m8MdiS2A
2dC2WGJLGG3NUG6cslxvIXP69hCn+Tu0TF5FDkKS6pwW4HmIYei23pbuqZzM
Jn13AQi25bovWQAP90j+3qOxFOj7HC8FmNGHEwN4Ffx9F5L4KbUT5st.Oj2n
EZj.fQqMTmnwj3AMj49PGOY0.6jb24tgpVu8oLhNicPhhImPuALNgB0AW6Zy
YIZoSJe.anJ8Jlut5ZHMptFhnZzq1KlCECcDTbYQLoTTYKsksECLYc6e7FVp
by9VZKJYUk10ZF+gT8kuihHHxWFZi6aoyMtGUn8fdkU.lWBgKasFu063PJZY
mjvq03qx37k6mpXdf.cc2N9tMeSoP7D08NDDB7IA8xNDzgwyaUoN9xNW9k34
8R77Npwy6ItHWN8rHWTXZDG8rkAimxhR3vWTGO7fpEgpsrNfwaNpuxESU4ja
cX9GXSrTesc5aoi6JtnPmakkyq9x6nxFVxTUZ4ph71SBWbCNObgVGV7Lm.3n
p3lqzpB1Q3FPni4Vq..NUEPACrHmmkVvdxb75Eq6VGwTzdgUf0ckOxzw4wrj
kV7BAxA0.clXSdZYbdvLYDhpU28fKgNwAywa1NCOxcesr1n2Y3OQMA5ae.T1
+0jl.eotQcWDtdSeaP2RoUTr4sAc6gUz2qd5zCvBpPEu53FnapXjJNCB+fIS
ll8MzEgiFK2PiHhZQDF7lq69F5JrABFKJWRqPQbtVCcsSpqpm+8vFv94NgZ.
ah4DxW56h4uuX2rR86IRxNWQdjoh5QANgOD20W.cO2jjUteGvn8MKwaCD+3J
+wmmjJfoGvVVl3i1cpIdTwkVJDCtdmFtbLL4iL9oUY4qe4HnBRtZ6H4V7B7g
WmN2c77h8xvEuvl8C2NDY+ZHhRm0aMmvCLw90o654+5LI9onaeHLNcGb9T39
QTu9rbGEJ2yzDaOK06UstmQznuRNF6.pmZ1wM0+qFw9eiJM8wpcGPfLWsyO6
1IP1Pz0NYVGQ5kXuPMa0qcx+9NZRV5CcCrMtz0rWQ7Ky8V6VOarPD467ra87
LSRyKU7IUgxs88rMMtbmCTIWWWN+IIQsw9PW+xNHmlriFlG18x9HynlWsmyo
MAC6b6hv8.+T7hrj282yx17We2+fKRS8E8hz4pd0taWZeVC6pp9V6YvW2vf3
JZZ0BfJGppgDadcXM1PQmthNjFgpvqTIOPjctdx8ppZs1cGMiMyW8Jb7IVTe
DqCoZGaop0qIV6yiXUfVqoA6I+bHdJ7fbFiGjiJ8nw3I0n3i1k8pdizCpudD
ZL5QXU5RG.vGv454dK4PcKcwA0RLwlsbfxH.okaSbpwnG0K1gNVOhbjdjGsd
OR5kGn5XJGBSIzyuSXqPmfLJL.J7fBFgmCREoAX6w3IohvXzXLJEoDqFYLdR
HUlf40OIXZ3CBuD9C4fvJ4fPJ40gSR2gRxggQhvh9tBejq+0q++.Dl+tU
-----------end_max5_patcher-----------
2 Likes
Thanks for this I’ll check that and update the fix. We knew that we were sacrificing the last slice but as this is a beginner’s patch, the coding poutine needed to add a slice was not worth in our mind the pedagogical cost… so I’ll probably fix the zl limits (a real bug) and add a note for the last item (instead of coding it since it is not in the film nor addressed anywhere)
this is where fluidbufcompose is your friend. You can ‘add’ an item at then end and we will resize under the hood for you. You could use fluid.list2buf to generate the 1 sample buffer you will want to copy at the end of the other.
keep it simple and add a note : sounds good to me
And thanks for the tip about bufcompose I’m not yet familiar with all the possibilities of this amazing lib.
Just for the record, and if it can help someone :
Here is a new version of the patch fixing the slices # issue by adding a “fake” slicepoint at the end of the buffer slicespoint
(so the number of points = number of slices, and all slices can be played in p playback)
----------begin_max5_patcher----------
4823.3oc6cs0aiqaD94r+JXMNOzKYSEunaEnn4.zVfBz1WZQ6CsEKjskSzYk
kbkjyY2sn42dI4PIK4HKSoHJ4z3DrwK0EqgeZ3vYFNyv+yGtYwxzuDlu.8qP
+CzM27e9vM2HOj3.2nZeyhsAeYUbPt7xVrNKc2ln3vE2BmLY+1nj3vB4YwpC
FsVdsoK+gOZW6BS2WTdkD0QgCU70cg.QrXws7+g9WpSuKnX0iQIO7orvUEvU
fIt2YcKhXK9K1AK9vleHz+RbK+2O7AwetUydypzsaCSJJowhvuHeJK9sYAOf
BPaRiWGlgVklTDDkvoCzSAwQqQe+e32+6QoYn+92+2PBvHG8XXVEjDGkDtJc
eh7qhpINwZGmrNMPP7oRfvSBDLYCar.HT2yx8KWFKwU7.fljvejSWkT0SAYI
AakeYK1DuOZ8cbPYUPQv90QoRH3EXX6WF590g4EQIAEQoInbNLsVWVI+1gH7
oXkVGsR7PBx9pNLTXlDHIDe3iAxQ0D1NBLN.CQI51qI8rWqUe0R1W8YlnqtC
8u2Gxwbc6ezwVBA04Ne4ONpe73CL7cD8UVsWt0+BByT8SUG8lEh2ROElkyYf
pQ32rHX2tZG9lZ2h.c9gT4Wj2sUGJJANjc0gxBeJp79spNZPFuaTv6C6yfQY
ewoTjf3qIkKGJYeT0fY46IEIIeiHFcluKXEbyhWbkm9.vX6Cu7kPgCyRhHdV
G.C9a7GhSW84v00ItEo6B4B+1kElykUJG313zqC2DrOt3Sa3hIyi9FHwgCws
c9MJJr0SVIg46yhBhq5.OjEsNMQPDMdSHNb4iiyYCyGXWuyHuhjfcsbybNON
rbhSly6j6yWFjIdQoDgRJOYQZZbySUcewgaJTmdWTRxQnXQ5tSexrnGdri6c
YJ+ja656Vdl7OsOAN6m37DEeJO3olncQPbrZXayu9uDjDsMnHrHBdEPrpNYX
R.ui9X9prz33F8W3LO0xYVy4wWE9iQqKdT9fpyLvu7nckLQKpdKuN5A9DCMO
VQvC4MORdwWAPu1g1uTMF9SEga2Ey6EMu.9nin7h7GS+wb0EVxnUG.NnFT8w
z0kA133GIKDjRUMf8Dx8NV1G9na3kS7eJYb1xAuTPbN0p7uU79R4BJkapCFR
BHYc3WpIIQI3WITYf.PiICNZBgrP9Hr0nG473wB979BS91mFmpeKmdtvSOeH
LouKH8vB2bBQyiMaxR2l+0sKSi6MnvLEn.7VJLwcxgjLNEGh9bRXPlPfvofE
R6vhm+ogERmvxs5CMDox1d1FAZ3yimG7PXqXSIpzPEK8wFpYYYvToBFddSMK
yx8a1Dl8LfJn6yC1tKGQ56HJO7.Yc1DmFTH3eVFj7v4k4.Jgp.KLyZxk4HsE
IlOmHgCbMMHqa9pSAbVCjuBduoK2kE7gq+7.XedcQVX3yn3zzOueWewHWWyK
WB6IkKgwTifPx93I62Vs2uIXyJygYconAzvTAjPFrNfJkWdkJARZABNXBkvW
VmPwXIEINe6XSd59rUkuLK0hC0j3pI4o1EJfDjtrt8lJXZREXCRDBkTzhHZA
y1FsdWZTRQdkM817w5fePwT2pVMtozr0fKNrl49CSy9CS5ihZMZu6fMS2gpY
2QHT2bLIVZREkjqQnhxt34oBeCREDrtTg0IoB0AK8a1BgyQV+IvQPeJnnHKZ
I2vi7icDXubXwCwoKChOxKCs4OiObf3FKut9OWrd+1cnjzrsAbk6BWiVGTDf
JRQ6hSKJBy9mKz1mrVioSmayerTlz2yXvsrWcH6UGxd0grWcH6H5P1y5fIgv
x95.EGKSZDGlQ.OBH0m2wex8tzPfDaiYeGVB.tRaYoyjM+hYPyCKdt1rp80.
OlQcHI.RXLn9Ov.cYX3O1yrLFfSNd6Z3O1e3F9qDT3O5PfQL72QWCYvlzDBa
cMgfYRpP9ki0fJbrLHUHFapEVXieGaN0lnBD3zW8sbB6ZbKmrHGlW7pgSWMb
5pgSWMbZ1ijkguJFp3W0lconN2vz10TJ6pBPkKlU4pK6GESXpggRmvRRrsYw
PvfZOq2hqdL1wjFQpFBhY5r5wFwDAQ+Su0FzjZEq6BTVRslgJzkHreGqZdoT
FUVDrh++SdnGZoyFUszgDRobIXeWGh4XpDEbs7upY9UMyupY9XpYdWpdsKKJ
o2wJK1xXZbAxDK0pmLyduOKb8dYtf0W8tLqZWXGvKFJ+YSm7E7gqvdQVPR9F
9zonfjf3ux4h0.sN0ZAYJ1IvvGGUJe5MObS62Fr6Yz8790Zt.rDw7qCHHiol
jihZUmgxPl5L0NH.5TDax6RGDnUHkqdu67FHYfFs0UEaYzgRkYFjkiHDDM0J
qpsv4JYxuBGr36YV4yXk.ZO57HfVgLeK74dmKUNlWprhSxeF7thutK2muIcr
g+kfOdz0EOlLjZ0M5hKesYF2Lo8hxaQu.7106WmcUpVt191h1Sea0PCiN7uE
ndgC48r6sn1GRvQOp7CWW2qN25pyst5bqWmysZVIlZJDDiV8X.+kRL5KHern
FLsupdCou8BrgaApCjMRXHr6TIAt0zmeyahC5ebFRmjrZ1F6UCibrmGS.D.E
GiJBSdFcOnrAZ6lUqDBjNpzSMHrDaR6DbqykQLTVz243LZ4.sb9HMWzeQLWE
GwV06AaCO1VovrpLGqYdrVEaS+cLwjLfi4K4SXkgBm0rMhShRMGuwaHGwcdP
7TkpByOPSgbD64df1eFsISn6YuKo.CdPFC6B4NuqIGjws6pPykbYWf.AJ3Zf
.pD0giTsMoiTANCJH7gvJ+6bH6YpVXgdNMO3bOEmi67MMu.hNHyQVGIQ2uIL
PX8WtRFDGEDCyxQ3Qb0JFMwO.DRMTpMM0EjLXvBlPeKDnhCZ0xoNFa0xAauc
HNyypkWfV1aoMTyKF1lUMy7zhGxh2aX1IwD6SfIcH0f0ElHYFUBOtEZpIB4A
oJJdxWbpf0PU6.8c3tsj5DClHtlZvjqSswR93YNxSNx6s5OAjs4mAxQsdcpB
p8zJvg+6p3vfr9K5A2krG1Yk8b6gOkDf1p9.yvA4cEd5074m7Qj0c8NBuHCc
MeA0BOOl.0uCU1nAkqmIlQp2UNS7fGbUx3n4.LhJpALCrzk3YtM2YERqJ4Bn
6OSiwJsnpnilhIBHwc5KwA7NCXr8fvEiU+LUiiT3hm8zKPNF8sXk7v95SzWS
ANn7gp2.JrCrKG3L03y2huKuHKLXKhft+aw7aSrvYHhEyq2bQ3IXpcBjjv3Y
zO7DQAFUhPGrRONZUnpXu0aTanwprfLzk8BrdkbIU+LwDSZnkPmF9ODKJ7iu
iIJrFFInlzMllv3KfvIxjgzDU63p5rESRBmevVXXfkp9VW1vTjttgBEkbNRG
HVaOxAJW1vTTNQWP2UOJ2AWixkMLUoTzczXWTDqMtFkC6.Qlgx0sFPZx3USx
xR0kkkzE7g8jFFn1Ld.UaZzPwMigHG0jryP2hLJbEphP6o6VSl3EnWg0ZRBK
iyzL2EcXou.ziJHFMjOI18KnOOm.Hhi6A9IYCiR4ZwO0BVeDsqp7oNpcyt5D
+zU3l6SOhoYOxuZ6TyV0ZJqcysQmcpSVGBxft.0mdf8R1vXrWzwhzU7SSHoq
qnEpQEsf801p.aiSG3K.5P6RZAwrzg1YafQqAfXcS2fVlvtc6Mn0l5Q13cYN
JrKN3qKCV8YsyQAb24nP6AB+K8wQdzCIAwcl7BDOoSlsAiTbdWWtwcgzwGhr
TaWWXKVv9Z5KbM8Etl9BuxxM99sKCyF6XcD+ZhCMsb.rpnEAhBrlgkUQDfdM
WqtYO7fAQI1fuJlbLob2xTtDt8dw3LVX0.fBChUjoeKXSr.kwSVj32C1DJ32
Vp2TG5CCJdP7MZzfPkqd.UE1dSdvf7CoQ8tj0X3wKTuYZ6KXHXAyrXAQsHqd
SNXD9kcYne52sAi9knuaC4mg94h7kwp2qAswB.XrJhOnpkgdxW09cgge9YsV
edZ+yxvQPxhBfvJVnoeeZdjXgbL7PrYiAJJYS5yPZXz2Yo6HbX70JwUjQ3ws
0NfN+GcCzZhOtAm2jWaaD9VYfH6PsdP4VEs2PiAUCgEqzk8+EoyRoM.z2.kT
qeABOY0ZacFxTJrlRlqfBNh+6xwr.AP0H8MTezigLk4R2atY6M7j8Swb8Cay
r1vlKfceSD5c5tHatuXQvNZQTfMATJj8.Lv5QWysJkZG+CLStg5pKQb1.GCB
VOlkJWCpZXHBW2vLjoGcSToe+gFybLZ5bNtUpkELUFDRSzCslx.bnWaRA8n+
Tu0TFBJZueraoGeEUMgwgFybY7yQO5F6Wa7frwbG.ndZJ.xql7GuYO5OwZR0
pBw4gFlJlCbGsf5AnU6Z72LCFts39HvuSAMX+5R7q2Z12IWNabdnnVrsaMZW
1xTzNSyHw8rQGISFUJXnvboDuKaYRJuGEW0yJQrtlMXCpYCloazZcVLGJPjJ
LujawjXNsm6YQmmS2u9nTSNQD9hHf9sz7cu+YGtAFy.QDVsFycA7US6HTDds
FuKCkMgWQhDY4sdQx1IJ1tjNK1t2p0FJEFLK19ccA2kAy6RrJ2vWIPz7QuFz
ZWCZsqAs1qLn0NiOzGwBRHQSGne9kiSUoQTIH2T657ArVKDrAWqEJrHKTrJa
al50ZorhEpyhILx6wV8qvEVBTtfQQVlYExEJ.sOZrCCTll09.wSGJ+KAa203
8v.1RWT3kxHxoO1PK4rfO+Khxsw8x9U+qliCthazOVLWlZrnpDQgeCwhoaYl
XLYwrI9GTv0TbXcUFdeHrPwdgjkyknDjnaEGNp074yDoANLfOAiMYs3sKXHX
8ZtEvRI3nfBTwigHQ.H+S+00vleV86TXkprFId7.HcfJugWZv8gRCtmWMjhR
mPjhqMFZaPxWAvJWvvHPKEN8aFMLhQGdoclBAfipzNCKFxnCRlIcGntlNcGJ
UeTYF4rMoVM0k9qb0+6c0WcZlPiVtTCPs8kN4QUHW.sXBGDLu+8h.8KLAY0a
3ZnQ1SUUYUW9JkeeMDR0UgsyZ.EzNhw1vVcTQfp2LkXQPDNcP0w9FfSDyFgS
J3oTSQq2pxgHSyVcgRQQU0vY5ECUE72+bQne+KEA98cWNIOP4XMXNML9MUre
iunC96RnUEV8L2IWrtv80E2hdYATTLb8VjX3ysHw59znhuVcCGj.BWa+0PkL
x46gtSkxfoRIJs7smgJXI2Rfco4gO2bei5U3oKiV3OKwMGP++oeWReDbEHaR
lNwARoixhql8ETBKLb69TqSF8UsKoSlxx+t72duDCc3mJ69T62a8yyFk1v7.
vzAT6oWK+sAYeNLKu9XryLw6ox6Ur4Jw2t0Ce+YXNyR2VMDbww2T3hccXYF2
3jRS3ZTH4e3SrsMrHKZExGcewibUMdLMdMx5N6i2Wk3BHj965UZt.iYz7w24
fTcl2kTIb1jUvYryg8HmKg7TaXy6QFdh54LBy6gu7WeF7zu9LJ8jza8YLRhF
Qc6WLtMuIrC0vDgNQoHwjUPVlU+R4AyjG.5VC+Xlrj95RzLJscwllJzgsvw2
zTASmHV+r4lA0AKC2.Gpr1PCETFYKSR6ZUNuOeZs4B08ZWKHyiHUsLIsORkK
V.mgk64PCSQ33wJQAvLYoQWstKNGZXpbfzernbEsBt9uVCSkDj3dlm.liJb5
QxKZlowzMOUbM511g1YKiqooBclAgZx5Cr1UDdyVX566Fox7lQOFkHzdpEqK
gx0rIyWesogSSD+eRNy7C4U101puz5xd1VitIcSZlymvLN24K+wQ8imnz6ox
XSR0GUeEhLkoBi.OisKNsPrm+Nl3Bte3hia2UE6SaAeqNFoLjTTapzvGtciB
+HmybSZ11mWLd4T05fUOymhgzIqS8UIytcXv5rE4aWv0yG6W9d1MhidJ7tGB
hRdtjNdJ3vHpWdVQRqHphsIgwUKmpUeGlzRpkYyZGGrOS0P+15+uFK6cisBu
tJV5dLvIpUq1Q0MAhtpjY0gzqmBh2Glt4kob1gtebZxCmFXaboa4uhDWlycV
sd1HoHxO5Z054yeLMqPymTIf112y9jnhpj4g8g5x4GjDUNvUjlLFxNX8U1QC
mKe5jPTE.Drx+NnQTmPvP0BFKWjg+bzpz3O9GSS28693eRHHR6Tvj0yoSZe0
haedEqx8xl1pboJXP9Mzzsm.pbrpFJr4klE0vTfSotWCcjegE79xD.xknp4Z
GZUcGMqKMu3U33Sr3yQrTURRypZ8Rh050QrZPq0LxavOGlqFOH5X7fn5ziFi
mTisNsS4VQ2Q5ActdDdL5QDc5RGAvGw455bG6XEMcfnSTEjXMaQgJkNqkaSd
pwnGcVrC2UOh0QOBVk5lsrY10jvHa8Z6DVZzIXiBCfFOH+Q34f0QZPcKmG9S
RGgw3wXTJVKVM1X7jv5LAyKeRvzvGUrCDOjiJxAGUfCdYwM3zE1fiKpARK5O
UwL3C+2O7+fhKsYH
-----------end_max5_patcher-----------
2 Likes
Thanks! I’ve been busting my head over this all day and this fixes it.
1 Like
here is an updated version with a cleaner & more straightforward usage of fluid.bufcompose~
----------begin_max5_patcher----------
4704.3oc6cs0aiiaE94jeErF6C6tMSVwK5VAJZVf1En.s6KKP6CcKFHaKmnY
jkbkjyjYVr42dI4gRVxQxh5lilYbBl3g5h0ge5vCO7vOd3uc8UKVF+je5Bze
B8ePWc0uc8UWIOj3.WoJe0hsdOsJzKUdYKVmDuaSPn+hafSFsea79rP+L4oI
piFrVdwwKe2aLyuR3xx93Ne34sXwM7+g9upSuyKa0CAQ2+1D+UYvUfI12ZbC
hXJ9K1BK9vjenh6g+vChxe1Xww98quV7mazr1rJd6V+nrbQLy+I4idweMw6d
jGZSb3Z+Dzp3nLufHtvgdzKLXM5G+6+zOghSP+6e7egDfQJ5A+jBHILHxeU7
9H4WEUcvk6WtLTVyw0hbFuD4XKZFZHtTIz3HgFlrfIdLglH+OvEhbQ3QujHu
sR4ewlv8AqukCJq7x71uNHVBAu.Cq+xP2s1OMKHxKKHNBkxgo00qJgeIf31n
pz5fUhuOujOpiBElIgMBwE9X5fsi.iCvPPj10ZRyMfzotZHqqtryVUcG5+s2
m+hPWCDDZ+sPPst0U9ik5GGdy.WKQck01K2xeq9Ip5ophd0BwaoG8SR4ZUkt
5qV3saWoCeUoaQfNuKV9E4bSwgBhfCYVbnD+GCxueihi5kvqaY7J19DnU1SV
4s+EeMwb6PQ6CJdUIeOoDI4aDQqyzcdqfaV7hK+zGPKSWPiPhOVLCIL4Xb.g
3uwuOLd068WWV3VDuyma7aWheJ2VorgakSu1ei29vr2tgalLM3SfQNNtW242
njvZOYgEleLIvKrnBbeRv53HgPT4Mg3v4ONt5NzIgY4Ji7Jh71UyMyU73vRC
mLkWI2mtzKQ7hRY0ljexr33vpmp39B82joN8tfnniPwr3cMexjf6e3D26xX9
I2dpua4YRe69H3rukqSj81TuGqh1YdggplsU+5exKJXqWleV.7JfXTbR+HOd
E8gzUIwggUpuvYdrlyrlqiux+CAqydP9fJqLvu7fc4JQKJdKuN3ddGCUOVl2
8oUORZ1GAPuzg1uT0F9sY9a2ExqEUu.dqifzrzGh+Pp5ByUzJC.GbCpba5x1
.qb7irEBltJZv1T26Ga6CW9Npw9lorMJELkSMx+agJdSV0.6BJmaJCFxmezZ
+mJYIQY3WYTom.PkNCNpCgDedKr0nG353gB87lgIb8vjqY4ao49AatuPnCea
vHgA93tFZnCwoGa1jDuM8iaWFG1cPgMLPAzsTXh87ARR3UEez6i78RDFDZDV
H0CKNtmFVtQengHcq1wrcngLhPCue7Tu68qEaxQkJtXouJiCcLTYvToeDNNm
Wb4DpLK2uYiexy.pftK0a6tTDo6pN3SBOaBi8xD5OK8htucaNfCnJvByLlO1
bjiEIj2mHgCbUGPVO0qLNIvAudzU6x.9v1clAXuecVhu+ynv332ueWmUtrsG
K6RXGocILlddQH42Zm0MH3wvlCyPWOfLds7.pmt.RHKzx4koxIv5LQeXHThX
Y0fiwRQUb95wlz38IqxeKm6DGppTWxxSoKTfHhKTq2McVJXZJE3ITHD9nnkP
TClsMX8t3fnrzhwzaxMB.AGESsKJU4lhSVCg3v3Ut9vzr9vjwnnTg5qN3oo5
P0r5HroOcJIFZJE4h6jHE4Uw1kB2ITJHXckBiFkB0Ayia1BQvQV+VHPPu0KK
KIXIefGoGGHvNEvh6CiW5EdTTFpKdFWeP3Fqnt9qKVue6NTTbxVOtyc9qQq8
x7PYwncgwYY9I+5B8i4rQ+B5bcwikxjwdFCgk8R.YuDP1KAj8R.Ym5.x1Z.l
DFK67f3rLF9f3vLBDQ.o+7VtymnKcZHogw2YNvw2gk.fsbrrT2YSHkfw7K5A
M0O64R8p1YcF1HDPR.jvXvKePAZtOveryXnX.A43KwA9icWnmgBW5Ya1eljA
9ao6.YvS4PHL0cHDroTJje4XMjBKiITJDMM0BKLweEObpMAYHHnucXjSX6Qb
jSFjC8MdYfSWF3zkANcYfSyTlrz1rXnXppIa1yjkd5s6vb1UQPk4+rbcpwOJ
5vTmAJ0DDZNFXHLfZGiY1HIG3rGisF9fHUMAwrAM6wSxPDDUO8lavozqXcmf
xbocZjBcEByuhcMO2JiZUDrh++ituKdoy5oW5vBRIe1VuPw7icLGSkPisg6E
Oyu3Y9EOyGSOyOkqW6RBh5gSqFCziKvdXtW8j4Zz6S7WuWt.w59D9LBtcgsf
fUnhm8Yl0ds3vdVhWT5Fd2oHuHuvOx0h0AsZZtfFl5DLvGK0x8zYloMseq2t
mQ2wEf0bCXQh9W6CIioCWihZTVgx4Lyr3IJ.APkhXR9pL.AZQob06cqYePBN
KyqJ1XDZJkuxfLrDLM7rOypZabtvl7PBvhqyXXeFqLP6PmYFnUHym7ety5Rt
VikUYklj6bJ5Jt5NcetSYfMbmCw3Q2P7LkTpUW1Em+ZaZByj1SJuAcFDsqud
C1Uta45GaqlymBU7x3Dw2BbwvhbI7VGEdKp4gU9nCU9gss8kfacI3VWBt0vB
tU0LwTUifXzpG73uTBQOgbwhbvz9h7MTG3aGqsQfZAK5HLP6N0h.2XFr.TyW
eyaB859BhGSGwU0rI1oDFYYNyVitBfhiQY9QOitCb1.scypUBCRGk5o5GVhG
93DrKqkQzYUzOlP3IamQyankxaoYi9EQeUbDaU2ar0F2VoPmmLKi4ZasB0ld
DXhQrAGyUpmvxY7lw7qEmDkp1diWP1hq+fnyX0PSgbDyyb7j0ug1Oi1jH78r
yMxraoQFCaCqcd6WkFY7wckMzIQXmm.Zx3tl.9JcLZzu.oBZFTv3Cgk+2Yks
mQdhE5X27Pv8TZN1yvt4EPzAaNx7HI5tM9dhQ+kprAwEPQyrTDtyfIYzL+.P
Hk59YwrzzVBICZrfIzY+rzL9yVN0ZfyVNLDaKh0La1xyPK6NXPWLRlgMYE8L
OSLxHSdu9IMiIrFvjSa0PpyoLdbCTTSDxAVpnZji1LOSsf7VCYsCz2faYjTM
kVfrGViIaqRskbwyUlmbTza6PGPliUOPVp4qSkLsmCMv3Fb3+tJz2K4TldXM
EPCMr8bygOkOGsc8A5gCVyU3YjmO+g2fLts6rNf3ngagsiIPt6PsRzfrxyLo
mKTX2GfdKMtxUbzrAFQwZ.MfkykCN7wbmjIGUI2.cOTZFXpEUwLZJlHfD6YT
JNf+PgAa2ObYf4OSU6HEtbtSrnmrcTH5SgJCkcNlnsmfCx+t0qAE1B1OCrlM
Fe+T3soYI9daQDzceJjeahINCQLXN8.uFst1IvBDFaYNy7+gO.chHAiJQnCi
ROLXkuJYu041dsDUUwSSW0KXzqD7Yl.S8j2fjgOPKgOM7eHFT3GWqyZh0XRH
0jtbZBimAzIZJozDUadU0ZxjjvUTLEiXvPk3qyKLUhttTghRZSzAg0zgbPxk
ElJImnKnaqmjagKI4xBSUpTzdzTWTBqItjjCaBQSijqaNfbJ4qlTkkpqJK4T
vG1QNhA0NzC3ySkBJsYLPozoTcFpVjQQqPkDZatZc1Lu.0JrVcRXL4JMu1Ic
XYn.zSJHSJkOIlcizmsY.hXYePeRVXRkbszmpAqOR1UoCUK0VbWYg+7k3l6R
MhoYMxsXOVyTU5bl6lqSNOoOYmvPFTEntzCpWxBSl5EcrDck9zYTz00zBcRM
sfc0dTAlStbfmAxg1ozBxzJGZuZClzb.HV2kaPMcXW+3Mnk55QV3qx0nvtPu
Otza060eMJfabMJjFbejW3gUoP8jhu97aiLPylv3QrtjtwOd0KXCqSefxol1
1vVrf4kkuvkkuvkkuv.S2362tzuGycBYPjcT6I0VkvhfV7FyooUQPPupyU24
hdvfECSHjDymzEW9tkobJb69rELPZ0.fBC3JxLZKXSLAkgiMS76fZBEBOK04
7NG+iNePbGA1fPkSR.UQau4CYPdWbP2Ya9nzdg5L219B5EVvFCrfnljUm4CX
3+ztDz29Mavne.8MaHeG56EKjFitabcfD.FqX7AUMMzymYsemu+6edXyOOcD
rrn.HrRERm8oY5mWpPViRSr4mBTPzl3mgkgQyXhaexZXEKbEICOtozAz4+nK
QqIt3JZdzYimOhXqzJx1qUwhJRKZugFCtFByIoM6KhkyR9X.l+KmkSng7GQ3
wNWaqSSlbi0TxriTvA7eOAuxo5jf.ZZ4an9nCMYxWKcyGSJCt29Qoy9W0956
4lY8nLbAr8m2TuS2IYy9ESB1QShBryfRgkU.CFVo8zMKkZy+A1Ttg5pqPzJw
w.x5wLTKBghBSjfqKMCY5I2D05x+PgWYNZZ0l1J0v.5iCnzD8PoyIAG5zlTP
GpOkKcNofh16G6F5oWQU8jbnvqbZ7yRO4F6Vp8frvqMAPczz.jSI6ONu5r+D
qoTqxPmGJLUbNvdzH0CHqlkzuYSHcawcwf+IMzfcKawuboW8cxkV44gRZwl1
kjcYooR1YZxD2VYGISxJELjwtTl2kklRIuCIW0VsHV1yF7D5YCloKasZEygD
DoByy0VlRLm1w8rn10zcK2JcJ6HBOKHzuglu6cas4FLXFfQXkJ7Zm.e0bbDJ
AuTguJoxlHpHAhk+80MGl05hDhFqh7lS3tXXDvlWR3tGSYMUzEwvLd3BqzdF
3l4ENqcgyZW3r1.3rV6IhN3yeQrV2uK0a6t9jJ0vliXhoyFr.Pwp7yBd9j8T
u2OSgWHYxAHHBIfrv9jphMZYdqrXPEGieUxrimBF7Vul6OkbxEPdYnrG7QB5
r8s+4RXy2U9NE97Hy3VGqCoET4zVhl0EF9isYIjhRmCHEuwMZqWzGAvJUnvH
PKEN8WFOLhQaKQgRgoyUknPgHnc9.odRdVp84g7rDUnhf8CgYD8YysRq0L40
DHNlVmo4wlBxRjz4S1yhabVL3DDzI1cBtj3GgL57bDSYsynTs1T.U5Tp.Ebt
QpSkhjL5SpQhLvs9OKEWlbzlh5mUlwcvOntiLiwbkqfmb2dzIKfetIF22KnE
2OHHE2smchUlq8.Vnw3uP3EGdFPLtbnUQ4Pl87wPkHbWURzc2fTIZpCMWuAA
ggYbYyptl2oF.3MirmUjet3dltKN0+4p6JFCwUhQHslkiaJ+Hrb9hxWK13uQ
rjmgXL+rfNlsMNDXR1HzoYOfkbNStsxeaFfLaX0nP5Tlss1OakCZVkhnJ0bF
444Vuj26mjVtMVacc1zp5AOzDXpcYxIpSudmqLXZdXT5EtX4NLbwrLrXOCy5
jwQo7mrP+g2w1V+rjfUHWzcYOj3m9Pb3ZjwslGuqQvMPHi+xfLsyFgUZn0AK
5Lmy7JP3UaSsFacH6+Oq2Tqmn88bU8mN+22yEAPYePOxbqCKLcUPBgL.Y3ZQ
fbVn2lUWk8yu4wVxLUWJmPmx7xC0ti7j5UcO.lMkzxloK4krmxb4pMQSxeYi
mZoPGkSK2oVJX5PDtVo7I0BKIyjEUlxIgEvtrzTJ6ZkkPams71P5zz1.HzLo
nzTJ6iTVnCvYXv7GJLUBNdr3eHlIy3ppnyacnvTszJbGKIWIqP.hKUXpVaE3
NR+voSJr5vZhXZ5LUa5uZOgXg1IW0oMGu10bR9qK4XMmC4bvl8t4KDNW9tzh
QOzNuJK3rvoFwPKCzVxHxh5BDifcgwYG1a2pea.+VW4OVpebDI3EE8+IEeTO
EMIiJtTSlSzxte3xovjOv0m1Dmr84SfJ37IrVs4EBeX2HUUGON4t1a0ybi1j
5gHiWBQllKZMIQZCA36kQ+b3uNCCdz+168BhJfyG8Nzh5kms1.CWo9vN0qbb
KIRyaJ++pLEfU1xTVHHeoZSItfoQZjrMcXP3pJhobC.YESWE1rNg0qG8B26G
u4kTV9.HDFGcey.akKcK+Uj3xrt0n1yFHMQ9FaiZOe5CwIYZ9jxeCT22y9nf
rB1fxttrc9dYQsxNrb61NX8y1Qkv30LI1UST7K1MkmLCCESImLLt+bvp3v27
Ohi282dy+TXRSeJ7yZjC+0Omb02qgQdZOu9reUyvf7JpFbI.UN1UCE17R+eq
34YS93TwwvWLTMWYlsyVsKrVpTwcTccM+hWgiuvhaSXopEYCqnzKEVigIrZH
qkhGTueNLaMdPzw3AQ0oFMFOoJa8FME+H6Q5A0VMBOF0HhNUoi.3izbsstkc
rukVD2RLroZIJjBNY0baxSMF0nVwN7opQrSTirMKWifRlLyRVXjkFZkvPiJA
aTT.z3A4NBOGrNVCHFiwSRGiw3wnUJVKUM1X7jv5zAyKeRP2vGsX4DOjiVjb
Gs.4d4hiq4EF2wKJN4H5aZwvc8ue8+Gv+NxWK
-----------end_max5_patcher-----------
1 Like
This was excellent @jamesbradbury . Really helpful to see the construction/deconstruction of bits that didn’t end up in the final patch too. Bravo!
2 Likes
Hi, I left a comment on the YT video not long ago asking if this could be done with a live audio input. So this is my idea of what I would like to achieve:
I want to map sound samples of the C3 note from a VST synth. In doing so, I would like to have mapped all the types of sounds found in this VSTs preset library. The problem arises when considering VST synth has around 14K presets, so saving individual sound files as a corpus is pretty much a dead end. I tried it, and it was a nightmare for the SDD.
My other idea was to do this in batches, but again, it’s pretty much a nightmare, and honestly, I don’t know if I can join different KMeans analyses into one.
Long story short, the only way is to have a constant audio source feeding into the mfccs analysis and mapping from there.
I’ve already made a MIDI loop that changes presets in the VST synth every time the loop finishes and restarts. But right now, my problem is I’m not able, or at least I don’t know if it’s possible to run parallel learning of the sound or if it should all be saved in a buffer and then analysed (I tried to dot the buffer as well but hit some walls int he middle).
Im open for suggestions … and help
1 Like