Hi Mihai,
OK I have just uploaded a package called SPEECHY.
Contains 1 MUX patch.
and A, E, I, O, U samples of my own voice.
I have tried the package on another PC and it seemed to work OK.
The samples are (0 - 127) in all 5 Multisamplas.
The samples are single and mapped a few notes above and below C3.
So just bang away on C3 while turning the knobs on the MUX patch play screen.
If you have any trouble mapping any vowel morph, such as "A" with "U" just let me know and I will send you the knobs values to turn etc.
Hopefully it will work on your PC and you can workout all the combinations yourself.
Anyway as I said this is one solution may be not the final solution but at least you can now morph 5 vowels.
This is more of a automation editing solution for a song score idea, rather than a real-time solution.
BUT
If you map a mod wheel to balancer [A E] you could play that in real-time and may be another controller on your keyboard to another balancer and you could do some real-time morphing.
Anyway here is the package, just put it in your /user/mux folder etc.
http://www.mediafire.com/file/6skci5o8u ... PEECHY.zip
Had a busy day a bit tired so excuse any typos etc.
OZ
The Choir Project Development/ Help
-
TheGuysanIdiot TheGuysanIdiot https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=213066
- KVRist
- 308 posts since 10 Aug, 2009 from United Kingdom
-
- KVRian
- Topic Starter
- 1233 posts since 29 Dec, 2008 from Lithuania
You are awesome.
I will test it and be back in one hour or so (now I'm on a rehearsal, we have sort of concert tomorrow).
I will test it and be back in one hour or so (now I'm on a rehearsal, we have sort of concert tomorrow).
-
- KVRian
- Topic Starter
- 1233 posts since 29 Dec, 2008 from Lithuania
A bit many buttons to use but quite a brilliant solution 
I think it is usable enough.
Soon all the choir will be up and working.
I think it is usable enough.
Soon all the choir will be up and working.
-
TheGuysanIdiot TheGuysanIdiot https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=213066
- KVRist
- 308 posts since 10 Aug, 2009 from United Kingdom
Hi Mihai,
Indeed it is quite complex a solution, but I have spent some time using it and once you get the hang of it I think it will work very well.
Anyway once you have done the choir samples I will try it in a short song demo thing.
There may be better or simpler solutions to follow.
But like most things in life lets see how practical it is.
Anyway off to bed soon as busy day again tomorrow.
Goodnight all.
OZ
Indeed it is quite complex a solution, but I have spent some time using it and once you get the hang of it I think it will work very well.
Anyway once you have done the choir samples I will try it in a short song demo thing.
There may be better or simpler solutions to follow.
But like most things in life lets see how practical it is.
Anyway off to bed soon as busy day again tomorrow.
Goodnight all.
OZ
-
- KVRian
- Topic Starter
- 1233 posts since 29 Dec, 2008 from Lithuania
Thanks OZ for all your work at this project. Greatly appreciated.
I will edit today more tenors just recorded and hopefully in the weekend I'll have more alto and soprano.
By next week we should have all the samples and I'll edit them (cut sounds, make loop points, normalize all at the same volume...).
I will edit today more tenors just recorded and hopefully in the weekend I'll have more alto and soprano.
By next week we should have all the samples and I'll edit them (cut sounds, make loop points, normalize all at the same volume...).
-
TheGuysanIdiot TheGuysanIdiot https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=213066
- KVRist
- 308 posts since 10 Aug, 2009 from United Kingdom
Hi Mihai,
I had odd dreams last night and gave me ideas on velocity.
The SPEECHY method I am sure will work once the user gets the idea and is very flexible but is complex and manual.
So I am going to think down another route, using velocity and mod wheels etc.
And then may be a hybrid of both SPEECHY and velocity etc.
Anyway the velocity idea was just during my dream so do not know how to do it yet.
I will post stuff during weekend etc if I come up with other solutions.
Bye for now have fun at concert.
OZ
I had odd dreams last night and gave me ideas on velocity.
The SPEECHY method I am sure will work once the user gets the idea and is very flexible but is complex and manual.
So I am going to think down another route, using velocity and mod wheels etc.
And then may be a hybrid of both SPEECHY and velocity etc.
Anyway the velocity idea was just during my dream so do not know how to do it yet.
I will post stuff during weekend etc if I come up with other solutions.
Bye for now have fun at concert.
OZ
-
TheGuysanIdiot TheGuysanIdiot https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=213066
- KVRist
- 308 posts since 10 Aug, 2009 from United Kingdom
Hi Mihai,
My dream last night is now becoming a reality.
The SPEECHY solution is mega complex but may be the way to go.
BUT
I am now working on a mega simple idea, looking good so far.
Must dash now will work on later tonight and post MUX patch and demo song, assuming it works OK.
No hints this time it will be a surprize.
Have a good day.
OZ
My dream last night is now becoming a reality.
The SPEECHY solution is mega complex but may be the way to go.
BUT
I am now working on a mega simple idea, looking good so far.
Must dash now will work on later tonight and post MUX patch and demo song, assuming it works OK.
No hints this time it will be a surprize.
Have a good day.
OZ
-
- KVRian
- Topic Starter
- 1233 posts since 29 Dec, 2008 from Lithuania
You like to cook us on steady fire 
Looking forward to your new patch solution. Meanwhile recording and editing more voices.
I plan for next year to get NOA Festival Choir, get a really really good concert hall for 2~3 days to record in and to make also attacks (LA, LE, LI, KA, KE, KI, MA, ME, MI...) and effects (glissando, clusters, shouts, whistling...etc).
That will be a bigger project, this is just the warm up.
Maybe I can get Jo to make a VSTi shell to that one.
Looking forward to your new patch solution. Meanwhile recording and editing more voices.
I plan for next year to get NOA Festival Choir, get a really really good concert hall for 2~3 days to record in and to make also attacks (LA, LE, LI, KA, KE, KI, MA, ME, MI...) and effects (glissando, clusters, shouts, whistling...etc).
That will be a bigger project, this is just the warm up.
Maybe I can get Jo to make a VSTi shell to that one.
-
- KVRian
- 855 posts since 3 Mar, 2009
Hi Mihai,
Downloaded your Test Choir.
Very nice work!
As a possible solution for a hard(er) transition between 2 samples (e.g. C3 - D3), you can try to use "Key Fader Type" in the sample properties, which fades in or fades out a sample over its key range.
I streched the AA-D3 sample from key 57-63 and Key Fade to "fade in" and the AA-C3 sample from key 59-63 using "fade out". The transition becomes smoother.
The added effect is that 2 samples play, which gives the sound more "choirus".
I still have to figure out how to fade in AND fade out a sample over its key range, but I have some ideas, which I will try out soon.
Andreas
Downloaded your Test Choir.
Very nice work!
As a possible solution for a hard(er) transition between 2 samples (e.g. C3 - D3), you can try to use "Key Fader Type" in the sample properties, which fades in or fades out a sample over its key range.
I streched the AA-D3 sample from key 57-63 and Key Fade to "fade in" and the AA-C3 sample from key 59-63 using "fade out". The transition becomes smoother.
The added effect is that 2 samples play, which gives the sound more "choirus".
I still have to figure out how to fade in AND fade out a sample over its key range, but I have some ideas, which I will try out soon.
Andreas
-
- KVRAF
- 7412 posts since 8 Feb, 2003 from London, UK
Same sample, two layers, one for the fade in up to mid point, one for the fade out. (MU.LAB's sample pool will only contain one copy of the sample -- but using multiple layers gives you more flexibility over how you set it up.)
-
TheGuysanIdiot TheGuysanIdiot https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=213066
- KVRist
- 308 posts since 10 Aug, 2009 from United Kingdom
Hi Mihai,
This is another prototype but this uses velocity mapping. So it is much different to previous complex solution.
This is a very simple idea but needs to be understood.
The MUX patch has:
1 balancer to set at "Single 0.0%" or "Multi 100%"
1 Multisampla that has only "A" loaded and uses (0 - 127)
1 Multisampla that has A, E, I, O, U samples each sample is set at one specific velocity
A = triggers only at 127
E = triggers only at 126
I = triggers only at 125
O =triggers only at 124
U =triggers only at 123
Here is how to use it:
Put the mux patch in slot 1 of rack 1
set the balancer to "Single" at 0.0%
Record some notes with MIDI keyboard you will hear "A" sample only.
Now set balancer at "Multi 100%"
Select all the notes you recorded and raise their velocity to 127 this will mean they will play "A" samples.
Now select all the notes you want to be "E" and set these notes at 126 velocity
and so on using the velocity trigger table above.
OK
What happens now it that any note set at the specific velocities will play the specific note.
As there is not much diffence in the volume of 127-123 all the notes A, E, I, O, U will sound at maximum loudless.
Now to get yours dynamic back map your mod wheel to the volume/gain slider of rack1 and record the automation.
Now edit the gain automation with a pen to put human feel back.
Important note:
When playing/writing your choir track set it to "Single" at 0.0% this way you will hear any note in the 0 - 127 velocity.
If you try playing via a MIDI keyboard using "Multi 100%" you will not hear anything below 122 velocity so it will seem like it is not working remember in this mode the velocity would need to be 124+. So you would need to hit keys hard.
All the above sounds hard to do when you type all this out and read it.
Try the MUX patch and you will see it is dead easy to use.
Again this is another route of discovery.
None of the solutions are perfect under all situations but one is complex the other is easier to understand and may make it more practical.
Please read all the above. I will post MUX patch and samples in an hour or so.
Give it a try and if you get stuck let me know and I will explain further if needed.
I actually think this method is probably the one I would use.
[more]
Your could always map the "Gain" of the AMP in the AEIOU Multisampla to your mod wheel to change the volume/gain there.
You could map the Release time of the same Multisampla to a controller so you could change the release for short vowel morphing etc.
PS it seems others are working on other solutions I will continue with my prototypes while others can present their ideas. So somewhere there will be a good solution among them for you to choose.
OZ
This is another prototype but this uses velocity mapping. So it is much different to previous complex solution.
This is a very simple idea but needs to be understood.
The MUX patch has:
1 balancer to set at "Single 0.0%" or "Multi 100%"
1 Multisampla that has only "A" loaded and uses (0 - 127)
1 Multisampla that has A, E, I, O, U samples each sample is set at one specific velocity
A = triggers only at 127
E = triggers only at 126
I = triggers only at 125
O =triggers only at 124
U =triggers only at 123
Here is how to use it:
Put the mux patch in slot 1 of rack 1
set the balancer to "Single" at 0.0%
Record some notes with MIDI keyboard you will hear "A" sample only.
Now set balancer at "Multi 100%"
Select all the notes you recorded and raise their velocity to 127 this will mean they will play "A" samples.
Now select all the notes you want to be "E" and set these notes at 126 velocity
and so on using the velocity trigger table above.
OK
What happens now it that any note set at the specific velocities will play the specific note.
As there is not much diffence in the volume of 127-123 all the notes A, E, I, O, U will sound at maximum loudless.
Now to get yours dynamic back map your mod wheel to the volume/gain slider of rack1 and record the automation.
Now edit the gain automation with a pen to put human feel back.
Important note:
When playing/writing your choir track set it to "Single" at 0.0% this way you will hear any note in the 0 - 127 velocity.
If you try playing via a MIDI keyboard using "Multi 100%" you will not hear anything below 122 velocity so it will seem like it is not working remember in this mode the velocity would need to be 124+. So you would need to hit keys hard.
All the above sounds hard to do when you type all this out and read it.
Try the MUX patch and you will see it is dead easy to use.
Again this is another route of discovery.
None of the solutions are perfect under all situations but one is complex the other is easier to understand and may make it more practical.
Please read all the above. I will post MUX patch and samples in an hour or so.
Give it a try and if you get stuck let me know and I will explain further if needed.
I actually think this method is probably the one I would use.
[more]
Your could always map the "Gain" of the AMP in the AEIOU Multisampla to your mod wheel to change the volume/gain there.
You could map the Release time of the same Multisampla to a controller so you could change the release for short vowel morphing etc.
PS it seems others are working on other solutions I will continue with my prototypes while others can present their ideas. So somewhere there will be a good solution among them for you to choose.
OZ
-
TheGuysanIdiot TheGuysanIdiot https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=213066
- KVRist
- 308 posts since 10 Aug, 2009 from United Kingdom
Hi Mihai,
The package has the same samples of my voice just put in /user/mux and load the MUX patch and search for files as usual.
This will be the last prototype for a while as I have stuff to do for most of the weekend it will also allow others to present their ideas from a different route.
Generally speaking all solutions presented by myself and others will work to some degree only time will tell if anyone can think of a perfect solution.
When you have done the final samples I will test them in my prototypes and also try the ideas presented by the others etc.
http://www.mediafire.com/file/uf8frh24j ... LOCITY.zip
OZ
The package has the same samples of my voice just put in /user/mux and load the MUX patch and search for files as usual.
This will be the last prototype for a while as I have stuff to do for most of the weekend it will also allow others to present their ideas from a different route.
Generally speaking all solutions presented by myself and others will work to some degree only time will tell if anyone can think of a perfect solution.
When you have done the final samples I will test them in my prototypes and also try the ideas presented by the others etc.
http://www.mediafire.com/file/uf8frh24j ... LOCITY.zip
OZ
-
TheGuysanIdiot TheGuysanIdiot https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=213066
- KVRist
- 308 posts since 10 Aug, 2009 from United Kingdom
Hi Mihai,
One last thing before I go to bed.
The long term solution for your planned large project probably would need programming code etc.
A choir VSTi is probably the way to go and that would morph all vowels and do more stuff etc.
All that is beyond me, indeed more Jo's area.
Goodnight all.
OZ
One last thing before I go to bed.
The long term solution for your planned large project probably would need programming code etc.
A choir VSTi is probably the way to go and that would morph all vowels and do more stuff etc.
All that is beyond me, indeed more Jo's area.
Goodnight all.
OZ
-
- KVRian
- Topic Starter
- 1233 posts since 29 Dec, 2008 from Lithuania
Thanks everyone for the input 
I like everyone's ideas. Somehow you guys are more imaginative than me.
So, for now I think the best idea would be to finish patches (I finished recordings) and make choir patch for each vowels and then let you fool around with them.
After all that is one of the best things, freedom. The reason I like Mu.Lab too
So, I'll try to make it as soon as I can.
I like everyone's ideas. Somehow you guys are more imaginative than me.
So, for now I think the best idea would be to finish patches (I finished recordings) and make choir patch for each vowels and then let you fool around with them.
After all that is one of the best things, freedom. The reason I like Mu.Lab too
So, I'll try to make it as soon as I can
.-
TheGuysanIdiot TheGuysanIdiot https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=213066
- KVRist
- 308 posts since 10 Aug, 2009 from United Kingdom
Morning Mihai,
Indeed it is probably best for you to finish the vowel patches then we will have something "real" to test morphing ideas with.
Ultimately the best solution will be the one that sounds most "real", morphs between the vowels smoothly and is also user friendly.
We may find that the final solution is a hybrid of all the solutions presented so far by others and myself etc.
I would also like to hear how Jo would do this as he has the best understanding of MU.LAB in general. But we will have to wait for his return from holidays.
Anyway I am finding all this very interesting as I am not a typical sample type user I tend to do most of my stuff with VSTi synths but I do use drum samples already. But a lot of VSTi these days have samples built-in.
So this is expanding my outlook in general.
One thing I know for sure is it is not easy to record samples, loop them and map them etc. All this is very time consuming so I think you are doing a great job making these sample packs for us.
Bye for now have a great weekend.
OZ
Indeed it is probably best for you to finish the vowel patches then we will have something "real" to test morphing ideas with.
Ultimately the best solution will be the one that sounds most "real", morphs between the vowels smoothly and is also user friendly.
We may find that the final solution is a hybrid of all the solutions presented so far by others and myself etc.
I would also like to hear how Jo would do this as he has the best understanding of MU.LAB in general. But we will have to wait for his return from holidays.
Anyway I am finding all this very interesting as I am not a typical sample type user I tend to do most of my stuff with VSTi synths but I do use drum samples already. But a lot of VSTi these days have samples built-in.
So this is expanding my outlook in general.
One thing I know for sure is it is not easy to record samples, loop them and map them etc. All this is very time consuming so I think you are doing a great job making these sample packs for us.
Bye for now have a great weekend.
OZ