Hello! I'm new to PD and was wondering if i could get a "working start" -
i'd like to someday be able to create a PD applet that can take audio input from a microphone and convert the sounds to text that can simultaneously be applied to a concurrently running program, like a word processor or a presentation program that accepts alpha-numeric input to trigger events. Can someone create a PD example that does the following:
- echo Alpha/numeric characters based on variable pitch
- echo alpha/numeric characters based on volume
- echo alpha/numeric characters based on rhythm (ie, number of pulses per
second or minute)
Sorry if this sounds simplistic - if i can see an example of a PD patch that can incorporate these elements, that will go a LONG way to helping me learn how to program PD on my own!
Thanks in advance for your help and suggestions, PD list!
speech to text in opensource software is still poor, and even poorer for languages other than english. afaik pd also doesnot have a builtin solution for that. you could try to use a different program for that and parse the text, or you could try more basic approaches within pd (like pitch recognition, amplitude/envelope following, or rhythm/melody recognition. I am sure there are some commercial solutions out there (like in the iphone) that have good speech-to-text recognition, including dictionaries. sorry that this probably is no real help for you, maybe someone else knows more? marius.
Hugh Sung wrote:
Hello! I'm new to PD and was wondering if i could get a "working start" -
i'd like to someday be able to create a PD applet that can take audio input from a microphone and convert the sounds to text that can simultaneously be applied to a concurrently running program, like a word processor or a presentation program that accepts alpha-numeric input to trigger events. Can someone create a PD example that does the following:
- echo Alpha/numeric characters based on variable pitch
- echo alpha/numeric characters based on volume
- echo alpha/numeric characters based on rhythm (ie, number of pulses
per second or minute)
Sorry if this sounds simplistic - if i can see an example of a PD patch that can incorporate these elements, that will go a LONG way to helping me learn how to program PD on my own!
Thanks in advance for your help and suggestions, PD list!
-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
marius schebella wrote:
speech to text in opensource software is still poor, and even poorer for languages other than english. afaik pd also doesnot have a builtin solution for that. you could try to use a different program for that and parse the text,
maybe sphinx (java) -> pdj -> pd (triggering events). would be a solution to try. marius.
Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals.
For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string
- the letter "A", perhaps.
I have a presentation program that can be configured to accept keyboard strokes to activate visual events.
Does that example make sense? PD sounds like it should be able to handle that capability pretty easily...
On Tue, Apr 29, 2008 at 12:14 PM, marius schebella < marius.schebella@gmail.com> wrote:
speech to text in opensource software is still poor, and even poorer for languages other than english. afaik pd also doesnot have a builtin solution for that. you could try to use a different program for that and parse the text, or you could try more basic approaches within pd (like pitch recognition, amplitude/envelope following, or rhythm/melody recognition. I am sure there are some commercial solutions out there (like in the iphone) that have good speech-to-text recognition, including dictionaries. sorry that this probably is no real help for you, maybe someone else knows more? marius.
Hugh Sung wrote:
Hello! I'm new to PD and was wondering if i could get a "working start"
i'd like to someday be able to create a PD applet that can take audio input from a microphone and convert the sounds to text that can simultaneously be applied to a concurrently running program, like a word processor or a presentation program that accepts alpha-numeric input to trigger events. Can someone create a PD example that does the following:
- echo Alpha/numeric characters based on variable pitch
- echo alpha/numeric characters based on volume
- echo alpha/numeric characters based on rhythm (ie, number of pulses
per second or minute)
Sorry if this sounds simplistic - if i can see an example of a PD patch that can incorporate these elements, that will go a LONG way to helping me learn how to program PD on my own!
Thanks in advance for your help and suggestions, PD list!
-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Hugh Sung wrote:
Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals.
For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps.
in that case, you best use fiddle~. what's the other program? does it accept network communication? then you can send the letters over tcp or udp connection (netsend). marius.
i'm not familiar with that program - can you provide a link to fiddle~, or any other program that can easily do what i'm looking for with all the various parameters?
Many thanks!
On Tue, Apr 29, 2008 at 12:50 PM, marius schebella < marius.schebella@gmail.com> wrote:
Hugh Sung wrote:
Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals.
For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps.
in that case, you best use fiddle~. what's the other program? does it accept network communication? then you can send the letters over tcp or udp connection (netsend). marius.
fiddle~ comes with pd (look in the "extra" folder), there is a help patch that will explain how it works. hope this helps. marius.
Hugh Sung wrote:
i'm not familiar with that program - can you provide a link to fiddle~, or any other program that can easily do what i'm looking for with all the various parameters?
Many thanks!
On Tue, Apr 29, 2008 at 12:50 PM, marius schebella <marius.schebella@gmail.com mailto:marius.schebella@gmail.com> wrote:
Hugh Sung wrote: Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals. For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps. in that case, you best use fiddle~. what's the other program? does it accept network communication? then you can send the letters over tcp or udp connection (netsend). marius.-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
As for forwarding keypresses and mouse events to an application ...
I have done this in Windows to control FLStudio (aka FruityLoops) by sending menu events and keypresses through python windows scripting, pywinauto, and osc in python.
I have info and code on my website. It should still work but it's a bit old (2006) and I don't use Windows anymore.
On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote:
fiddle~ comes with pd (look in the "extra" folder), there is a help patch that will explain how it works. hope this helps. marius.
Hugh Sung wrote:
i'm not familiar with that program - can you provide a link to fiddle~, or any other program that can easily do what i'm looking for with all the various parameters?
Many thanks!
On Tue, Apr 29, 2008 at 12:50 PM, marius schebella <marius.schebella@gmail.com mailto:marius.schebella@gmail.com> wrote:
Hugh Sung wrote: Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals. For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps. in that case, you best use fiddle~. what's the other program? does it accept network communication? then you can send the letters over tcp or udp connection (netsend). marius.-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
Dan Wilcox robotcowboy.com
I'm using a presentation program called "Liquid Media", sort of like Powerpoint on steroids. Up to this point i had been using multiple foot pedals assigned to different keystrokes to trigger visual cues, like video clips, animated GIFs and the like, during live classical music piano recitals.
I'd like to incorporate the dynamic capabilities of realtime visual sync, with the ability to easily "change the scene", as it were. i'm hoping that someone can help me come up with a combined solution where PD acts as an audio trigger controller, triggering keystroke events based on pitch ranges, dynamic peaks and rhythmic events, and Liquid Media accepts those PD-generated keystrokes to trigger correlating visual cues. Each scene would comprise of specific trigger-able events, and the foot switch would allow for a manual "change of scene". i'm hoping to accomplish two things with this combined approach:
- make it a little easier to incorporate visuals synced to audio events
- allow for easy scene transitions, to better compliment the "narrative"
structure of classical compositions
Hope this makes sense...thanks all for your input so far! (and patience!)
On Tue, Apr 29, 2008 at 2:45 PM, Dan Wilcox danomatika@gmail.com wrote:
As for forwarding keypresses and mouse events to an application ...
I have done this in Windows to control FLStudio (aka FruityLoops) by sending menu events and keypresses through python windows scripting, pywinauto, and osc in python.
I have info and code on my website http://www.robotcowboy.com/?p=60. It should still work but it's a bit old (2006) and I don't use Windows anymore.
On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote:
fiddle~ comes with pd (look in the "extra" folder), there is a help patch that will explain how it works. hope this helps. marius.
Hugh Sung wrote:
i'm not familiar with that program - can you provide a link to fiddle~, or any other program that can easily do what i'm looking for with all the various parameters?
Many thanks!
On Tue, Apr 29, 2008 at 12:50 PM, marius schebella <marius.schebella@gmail.com <mailto:marius.schebella@gmail.com marius.schebella@gmail.com>> wrote:
Hugh Sung wrote: Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals. For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps. in that case, you best use fiddle~. what's the other program? does it accept network communication? then you can send the letters over tcp or udp connection (netsend). marius.-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
Dan Wilcox robotcowboy.com http://www.robotcowboy.com
The [fiddle~] and [bonk~] objects should let you capture all of the information that you are looking for. As far as getting that information from Pd to Liquid Media, that may be the tricky part. Your current pedals are sending keystrokes straight to Liquid Media. Do you know if there is a way to control Liquid Media via MIDI messages, or some sort of message sent via a socket connection (maybe OSC)? If so, then you might be able to use Pd to capture your pedal input and then combine that with the information from [fiddle~] and/or [bonk~] to send the appropriate information to Liquid Media. If the only way to control Liquid Media is through keystrokes, then you might need to write some software that intercepts the keystrokes from the pedals and changes them depending on what is going on in Pd.
andy
On Tue, Apr 29, 2008 at 12:18 PM, Hugh Sung hughsung@gmail.com wrote:
I'm using a presentation program called "Liquid Media", sort of like Powerpoint on steroids. Up to this point i had been using multiple foot pedals assigned to different keystrokes to trigger visual cues, like video clips, animated GIFs and the like, during live classical music piano recitals.
I'd like to incorporate the dynamic capabilities of realtime visual sync, with the ability to easily "change the scene", as it were. i'm hoping that someone can help me come up with a combined solution where PD acts as an audio trigger controller, triggering keystroke events based on pitch ranges, dynamic peaks and rhythmic events, and Liquid Media accepts those PD-generated keystrokes to trigger correlating visual cues. Each scene would comprise of specific trigger-able events, and the foot switch would allow for a manual "change of scene". i'm hoping to accomplish two things with this combined approach:
- make it a little easier to incorporate visuals synced to audio events
- allow for easy scene transitions, to better compliment the "narrative"
structure of classical compositions
Hope this makes sense...thanks all for your input so far! (and patience!)
On Tue, Apr 29, 2008 at 2:45 PM, Dan Wilcox danomatika@gmail.com wrote:
As for forwarding keypresses and mouse events to an application ...
I have done this in Windows to control FLStudio (aka FruityLoops) by
sending menu events and keypresses through python windows scripting, pywinauto, and osc in python.
I have info and code on my website. It should still work but it's a bit
old (2006) and I don't use Windows anymore.
On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote: fiddle~ comes with pd (look in the "extra" folder), there is a help patch that will explain how it works. hope this helps. marius.
Hugh Sung wrote:
i'm not familiar with that program - can you provide a link to fiddle~, or any other program that can easily do what i'm looking for with all the various parameters?
Many thanks!
On Tue, Apr 29, 2008 at 12:50 PM, marius schebella <marius.schebella@gmail.com mailto:marius.schebella@gmail.com> wrote:
Hugh Sung wrote:
Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals.
For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps.
in that case, you best use fiddle~. what's the other program? does it accept network communication? then you can send the letters over tcp or udp connection (netsend). marius.
-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
Dan Wilcox robotcowboy.com
-- Hugh Sung www.hughsung.com
www.TabletPCMusician.com www.musicmeetstech.com _______________________________________________ PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
The pedals basically operate as programmable usb keyboard extensions, where you can custom assign any keystroke to any pedal (up to 12). I'm hoping that by assigning a certain set of keys to the pedals and reserving the rest for PD to control that everything can simply work together, if PD can output keyboard letters/numbers. Everything would run from the same computer. Again, please forgive me if i sound completely ignorant regarding what's possible with PD... Many thanks again!
On Tue, Apr 29, 2008 at 6:13 PM, Andrew Turley aturley@acm.org wrote:
The [fiddle~] and [bonk~] objects should let you capture all of the information that you are looking for. As far as getting that information from Pd to Liquid Media, that may be the tricky part. Your current pedals are sending keystrokes straight to Liquid Media. Do you know if there is a way to control Liquid Media via MIDI messages, or some sort of message sent via a socket connection (maybe OSC)? If so, then you might be able to use Pd to capture your pedal input and then combine that with the information from [fiddle~] and/or [bonk~] to send the appropriate information to Liquid Media. If the only way to control Liquid Media is through keystrokes, then you might need to write some software that intercepts the keystrokes from the pedals and changes them depending on what is going on in Pd.
andy
On Tue, Apr 29, 2008 at 12:18 PM, Hugh Sung hughsung@gmail.com wrote:
I'm using a presentation program called "Liquid Media", sort of like Powerpoint on steroids. Up to this point i had been using multiple foot pedals assigned to different keystrokes to trigger visual cues, like
video
clips, animated GIFs and the like, during live classical music piano recitals.
I'd like to incorporate the dynamic capabilities of realtime visual
sync,
with the ability to easily "change the scene", as it were. i'm hoping
that
someone can help me come up with a combined solution where PD acts as an audio trigger controller, triggering keystroke events based on pitch
ranges,
dynamic peaks and rhythmic events, and Liquid Media accepts those PD-generated keystrokes to trigger correlating visual cues. Each scene would comprise of specific trigger-able events, and the foot switch
would
allow for a manual "change of scene". i'm hoping to accomplish two
things
with this combined approach:
- make it a little easier to incorporate visuals synced to audio
events
- allow for easy scene transitions, to better compliment the
"narrative"
structure of classical compositions
Hope this makes sense...thanks all for your input so far! (and
patience!)
On Tue, Apr 29, 2008 at 2:45 PM, Dan Wilcox danomatika@gmail.com
wrote:
As for forwarding keypresses and mouse events to an application ...
I have done this in Windows to control FLStudio (aka FruityLoops) by
sending menu events and keypresses through python windows scripting, pywinauto, and osc in python.
I have info and code on my website. It should still work but it's a
bit
old (2006) and I don't use Windows anymore.
On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote: fiddle~ comes with pd (look in the "extra" folder), there is a help patch that will explain how it works. hope this helps. marius.
Hugh Sung wrote:
i'm not familiar with that program - can you provide a link to
fiddle~,
or any other program that can easily do what i'm looking for with
all
the various parameters?
Many thanks!
On Tue, Apr 29, 2008 at 12:50 PM, marius schebella <marius.schebella@gmail.com mailto:marius.schebella@gmail.com>
wrote:
Hugh Sung wrote:
Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals.
For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps.
in that case, you best use fiddle~. what's the other program? does it accept network communication? then you can send the letters over tcp or udp connection (netsend). marius.
-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
Dan Wilcox robotcowboy.com
-- Hugh Sung www.hughsung.com
www.TabletPCMusician.com www.musicmeetstech.com _______________________________________________ PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Here's one scenario i'd like to see:
- Pedal assigned [PgDn] keystroke within Liquid Media, which triggers
scene change 2. PD outputs letters A-Z, depending on either volume or pitch frequency ranges input from audio source (microphone into acoustic piano) 3. A given scene within Liquid Media would then interpret the PD-generated keystrokes and trigger image events based on letter key "depressed" by PD.
You'd have the ease of image generation (a la Powerpoint) with the speed and power of PD to map the sound of acoustic music. Just my newbie imagination going here.
On Tue, Apr 29, 2008 at 6:46 PM, Hugh Sung hughsung@gmail.com wrote:
The pedals basically operate as programmable usb keyboard extensions, where you can custom assign any keystroke to any pedal (up to 12). I'm hoping that by assigning a certain set of keys to the pedals and reserving the rest for PD to control that everything can simply work together, if PD can output keyboard letters/numbers. Everything would run from the same computer. Again, please forgive me if i sound completely ignorant regarding what's possible with PD... Many thanks again!
On Tue, Apr 29, 2008 at 6:13 PM, Andrew Turley aturley@acm.org wrote:
The [fiddle~] and [bonk~] objects should let you capture all of the information that you are looking for. As far as getting that information from Pd to Liquid Media, that may be the tricky part. Your current pedals are sending keystrokes straight to Liquid Media. Do you know if there is a way to control Liquid Media via MIDI messages, or some sort of message sent via a socket connection (maybe OSC)? If so, then you might be able to use Pd to capture your pedal input and then combine that with the information from [fiddle~] and/or [bonk~] to send the appropriate information to Liquid Media. If the only way to control Liquid Media is through keystrokes, then you might need to write some software that intercepts the keystrokes from the pedals and changes them depending on what is going on in Pd.
andy
On Tue, Apr 29, 2008 at 12:18 PM, Hugh Sung hughsung@gmail.com wrote:
I'm using a presentation program called "Liquid Media", sort of like Powerpoint on steroids. Up to this point i had been using multiple
foot
pedals assigned to different keystrokes to trigger visual cues, like
video
clips, animated GIFs and the like, during live classical music piano recitals.
I'd like to incorporate the dynamic capabilities of realtime visual
sync,
with the ability to easily "change the scene", as it were. i'm hoping
that
someone can help me come up with a combined solution where PD acts as
an
audio trigger controller, triggering keystroke events based on pitch
ranges,
dynamic peaks and rhythmic events, and Liquid Media accepts those PD-generated keystrokes to trigger correlating visual cues. Each
scene
would comprise of specific trigger-able events, and the foot switch
would
allow for a manual "change of scene". i'm hoping to accomplish two
things
with this combined approach:
- make it a little easier to incorporate visuals synced to audio
events
- allow for easy scene transitions, to better compliment the
"narrative"
structure of classical compositions
Hope this makes sense...thanks all for your input so far! (and
patience!)
On Tue, Apr 29, 2008 at 2:45 PM, Dan Wilcox danomatika@gmail.com
wrote:
As for forwarding keypresses and mouse events to an application ...
I have done this in Windows to control FLStudio (aka FruityLoops) by
sending menu events and keypresses through python windows scripting, pywinauto, and osc in python.
I have info and code on my website. It should still work but it's a
bit
old (2006) and I don't use Windows anymore.
On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote: fiddle~ comes with pd (look in the "extra" folder), there is a help patch that will explain how it works. hope this helps. marius.
Hugh Sung wrote:
i'm not familiar with that program - can you provide a link to
fiddle~,
or any other program that can easily do what i'm looking for with
all
the various parameters?
Many thanks!
On Tue, Apr 29, 2008 at 12:50 PM, marius schebella <marius.schebella@gmail.com mailto:marius.schebella@gmail.com>
wrote:
Hugh Sung wrote:
Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals.
For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps.
in that case, you best use fiddle~. what's the other program? does it accept network communication? then you can send the letters
over
tcp or udp connection (netsend). marius.
-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
Dan Wilcox robotcowboy.com
-- Hugh Sung www.hughsung.com
www.TabletPCMusician.com www.musicmeetstech.com _______________________________________________ PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-- Hugh Sung www.hughsung.com www.TabletPCMusician.com www.musicmeetstech.com
Le Tue, 29 Apr 2008 18:50:33 -0400, "Hugh Sung" hughsung@gmail.com a écrit :
Python might bring you the right solution:
http://www.rutherfurd.net/python/sendkeys/
You can either send your letters/keystrokes from inside pd using the pyext external that loads your python script
or send messages via OSC to a python script that implements the ixi software simpleOSC lib
http://www.ixi-software.net/content/backyard.html
++ O.
Here's one scenario i'd like to see:
- Pedal assigned [PgDn] keystroke within Liquid Media, which triggers
scene change 2. PD outputs letters A-Z, depending on either volume or pitch frequency ranges input from audio source (microphone into acoustic piano) 3. A given scene within Liquid Media would then interpret the PD-generated keystrokes and trigger image events based on letter key "depressed" by PD.
You'd have the ease of image generation (a la Powerpoint) with the speed and power of PD to map the sound of acoustic music. Just my newbie imagination going here.
On Tue, Apr 29, 2008 at 6:46 PM, Hugh Sung hughsung@gmail.com wrote:
The pedals basically operate as programmable usb keyboard extensions, where you can custom assign any keystroke to any pedal (up to 12). I'm hoping that by assigning a certain set of keys to the pedals and reserving the rest for PD to control that everything can simply work together, if PD can output keyboard letters/numbers. Everything would run from the same computer. Again, please forgive me if i sound completely ignorant regarding what's possible with PD... Many thanks again!
On Tue, Apr 29, 2008 at 6:13 PM, Andrew Turley aturley@acm.org wrote:
The [fiddle~] and [bonk~] objects should let you capture all of the information that you are looking for. As far as getting that information from Pd to Liquid Media, that may be the tricky part. Your current pedals are sending keystrokes straight to Liquid Media. Do you know if there is a way to control Liquid Media via MIDI messages, or some sort of message sent via a socket connection (maybe OSC)? If so, then you might be able to use Pd to capture your pedal input and then combine that with the information from [fiddle~] and/or [bonk~] to send the appropriate information to Liquid Media. If the only way to control Liquid Media is through keystrokes, then you might need to write some software that intercepts the keystrokes from the pedals and changes them depending on what is going on in Pd.
andy
On Tue, Apr 29, 2008 at 12:18 PM, Hugh Sung hughsung@gmail.com wrote:
I'm using a presentation program called "Liquid Media", sort of like Powerpoint on steroids. Up to this point i had been using multiple
foot
pedals assigned to different keystrokes to trigger visual cues, like
video
clips, animated GIFs and the like, during live classical music piano recitals.
I'd like to incorporate the dynamic capabilities of realtime visual
sync,
with the ability to easily "change the scene", as it were. i'm hoping
that
someone can help me come up with a combined solution where PD acts as
an
audio trigger controller, triggering keystroke events based on pitch
ranges,
dynamic peaks and rhythmic events, and Liquid Media accepts those PD-generated keystrokes to trigger correlating visual cues. Each
scene
would comprise of specific trigger-able events, and the foot switch
would
allow for a manual "change of scene". i'm hoping to accomplish two
things
with this combined approach:
- make it a little easier to incorporate visuals synced to audio
events
- allow for easy scene transitions, to better compliment the
"narrative"
structure of classical compositions
Hope this makes sense...thanks all for your input so far! (and
patience!)
On Tue, Apr 29, 2008 at 2:45 PM, Dan Wilcox danomatika@gmail.com
wrote:
As for forwarding keypresses and mouse events to an application ...
I have done this in Windows to control FLStudio (aka FruityLoops) by
sending menu events and keypresses through python windows scripting, pywinauto, and osc in python.
I have info and code on my website. It should still work but it's a
bit
old (2006) and I don't use Windows anymore.
On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote: fiddle~ comes with pd (look in the "extra" folder), there is a help patch that will explain how it works. hope this helps. marius.
Hugh Sung wrote:
i'm not familiar with that program - can you provide a link to
fiddle~,
or any other program that can easily do what i'm looking for with
all
the various parameters?
Many thanks!
On Tue, Apr 29, 2008 at 12:50 PM, marius schebella <marius.schebella@gmail.com mailto:marius.schebella@gmail.com>
wrote:
Hugh Sung wrote:
Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals.
For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps.
in that case, you best use fiddle~. what's the other program? does it accept network communication? then you can send the letters
over
tcp or udp connection (netsend). marius.
-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
Dan Wilcox robotcowboy.com
-- Hugh Sung www.hughsung.com
www.TabletPCMusician.com www.musicmeetstech.com _______________________________________________ PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-- Hugh Sung www.hughsung.com www.TabletPCMusician.com www.musicmeetstech.com
We're having a neat discussion about this in the Pure Data forum now: http://puredata.hurleur.com/sujet-1686-fiddle-piano
jon2211 came up with a neat patch that demonstrates how well fiddle~ recognizes notes played on a piano within a given range, triggering a key text output. Another poster showed an example of modifying that to trigger with envelope/volume changes. Of course, we're still stuck with how to port the key stroke messages to an external application.
Would someone be willing to write up a python script that can intercept Pure Data text echoes and forward them to other concurrently running applications, like PowerPoint or Word? This is the missing link in the puzzle - heck, i'd even be willing to pay someone if they could come up with a useful workaround!
Keeping my fingers crossed... Many thanks as always!
On Wed, Apr 30, 2008 at 6:11 AM, Olivier Heinry olivier@heinry.fr wrote:
Le Tue, 29 Apr 2008 18:50:33 -0400, "Hugh Sung" hughsung@gmail.com a écrit :
Python might bring you the right solution:
http://www.rutherfurd.net/python/sendkeys/
You can either send your letters/keystrokes from inside pd using the pyext external that loads your python script
or send messages via OSC to a python script that implements the ixi software simpleOSC lib
http://www.ixi-software.net/content/backyard.html
++ O.
Here's one scenario i'd like to see:
- Pedal assigned [PgDn] keystroke within Liquid Media, which triggers
scene change 2. PD outputs letters A-Z, depending on either volume or pitch
frequency
ranges input from audio source (microphone into acoustic piano) 3. A given scene within Liquid Media would then interpret the
PD-generated
keystrokes and trigger image events based on letter key "depressed" by
PD.
You'd have the ease of image generation (a la Powerpoint) with the speed
and
power of PD to map the sound of acoustic music. Just my newbie
imagination
going here.
On Tue, Apr 29, 2008 at 6:46 PM, Hugh Sung hughsung@gmail.com wrote:
The pedals basically operate as programmable usb keyboard extensions, where you can custom assign any keystroke to any pedal (up to 12).
I'm
hoping that by assigning a certain set of keys to the pedals and
reserving
the rest for PD to control that everything can simply work together,
if PD
can output keyboard letters/numbers. Everything would run from the
same
computer. Again, please forgive me if i sound completely ignorant
regarding
what's possible with PD... Many thanks again!
On Tue, Apr 29, 2008 at 6:13 PM, Andrew Turley aturley@acm.org
wrote:
The [fiddle~] and [bonk~] objects should let you capture all of the information that you are looking for. As far as getting that information from Pd to Liquid Media, that may be the tricky part.
Your
current pedals are sending keystrokes straight to Liquid Media. Do
you
know if there is a way to control Liquid Media via MIDI messages, or some sort of message sent via a socket connection (maybe OSC)? If
so,
then you might be able to use Pd to capture your pedal input and
then
combine that with the information from [fiddle~] and/or [bonk~] to send the appropriate information to Liquid Media. If the only way to control Liquid Media is through keystrokes, then you might need to write some software that intercepts the keystrokes from the pedals
and
changes them depending on what is going on in Pd.
andy
On Tue, Apr 29, 2008 at 12:18 PM, Hugh Sung hughsung@gmail.com
wrote:
I'm using a presentation program called "Liquid Media", sort of
like
Powerpoint on steroids. Up to this point i had been using
multiple
foot
pedals assigned to different keystrokes to trigger visual cues,
like
video
clips, animated GIFs and the like, during live classical music
piano
recitals.
I'd like to incorporate the dynamic capabilities of realtime
visual
sync,
with the ability to easily "change the scene", as it were. i'm
hoping
that
someone can help me come up with a combined solution where PD acts
as
an
audio trigger controller, triggering keystroke events based on
pitch
ranges,
dynamic peaks and rhythmic events, and Liquid Media accepts those PD-generated keystrokes to trigger correlating visual cues. Each
scene
would comprise of specific trigger-able events, and the foot
switch
would
allow for a manual "change of scene". i'm hoping to accomplish
two
things
with this combined approach:
- make it a little easier to incorporate visuals synced to audio
events
- allow for easy scene transitions, to better compliment the
"narrative"
structure of classical compositions
Hope this makes sense...thanks all for your input so far! (and
patience!)
On Tue, Apr 29, 2008 at 2:45 PM, Dan Wilcox danomatika@gmail.com
wrote:
As for forwarding keypresses and mouse events to an application
...
I have done this in Windows to control FLStudio (aka
FruityLoops) by
sending menu events and keypresses through python windows
scripting,
pywinauto, and osc in python.
I have info and code on my website. It should still work but
it's a
bit
old (2006) and I don't use Windows anymore.
On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote: fiddle~ comes with pd (look in the "extra" folder), there is a
help
patch that will explain how it works. hope this helps. marius.
Hugh Sung wrote: > i'm not familiar with that program - can you provide a link to
fiddle~,
> or any other program that can easily do what i'm looking for
with
all
> the various parameters? > > Many thanks! > > On Tue, Apr 29, 2008 at 12:50 PM, marius schebella > <marius.schebella@gmail.com <mailto:marius.schebella@gmail.com
wrote:
> > Hugh Sung wrote: > > Actually, this isn't for speech recognition - i'm trying to
come
> up with a simple way to use key triggers to activate visuals. > > For example: if the piano plays A440, i'm presuming PD can be > configured to recognize that pitch with a microphone input and > then echo out a text string - the letter "A", perhaps. > > > in that case, you best use fiddle~. what's the other program?
does
> it accept network communication? then you can send the letters
over
> tcp or udp connection (netsend). > marius. > > > > > -- > Hugh Sung > www.hughsung.com http://www.hughsung.com > www.TabletPCMusician.com http://www.TabletPCMusician.com > www.musicmeetstech.com http://www.musicmeetstech.com
Dan Wilcox robotcowboy.com
-- Hugh Sung www.hughsung.com
www.TabletPCMusician.com www.musicmeetstech.com _______________________________________________ PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-- Hugh Sung www.hughsung.com www.TabletPCMusician.com www.musicmeetstech.com
Write them to a text file then import them.
On Tue, 6 May 2008 22:54:54 -0400 "Hugh Sung" hughsung@gmail.com wrote:
We're having a neat discussion about this in the Pure Data forum now: http://puredata.hurleur.com/sujet-1686-fiddle-piano
jon2211 came up with a neat patch that demonstrates how well fiddle~ recognizes notes played on a piano within a given range, triggering a key text output. Another poster showed an example of modifying that to trigger with envelope/volume changes. Of course, we're still stuck with how to port the key stroke messages to an external application.
Would someone be willing to write up a python script that can intercept Pure Data text echoes and forward them to other concurrently running applications, like PowerPoint or Word? This is the missing link in the puzzle - heck, i'd even be willing to pay someone if they could come up with a useful workaround!
Keeping my fingers crossed... Many thanks as always!
On Wed, Apr 30, 2008 at 6:11 AM, Olivier Heinry olivier@heinry.fr wrote:
Le Tue, 29 Apr 2008 18:50:33 -0400, "Hugh Sung" hughsung@gmail.com a écrit :
Python might bring you the right solution:
http://www.rutherfurd.net/python/sendkeys/
You can either send your letters/keystrokes from inside pd using the pyext external that loads your python script
or send messages via OSC to a python script that implements the ixi software simpleOSC lib
http://www.ixi-software.net/content/backyard.html
++ O.
Here's one scenario i'd like to see:
- Pedal assigned [PgDn] keystroke within Liquid Media, which triggers
scene change 2. PD outputs letters A-Z, depending on either volume or pitch
frequency
ranges input from audio source (microphone into acoustic piano) 3. A given scene within Liquid Media would then interpret the
PD-generated
keystrokes and trigger image events based on letter key "depressed" by
PD.
You'd have the ease of image generation (a la Powerpoint) with the speed
and
power of PD to map the sound of acoustic music. Just my newbie
imagination
going here.
On Tue, Apr 29, 2008 at 6:46 PM, Hugh Sung hughsung@gmail.com wrote:
The pedals basically operate as programmable usb keyboard extensions, where you can custom assign any keystroke to any pedal (up to 12).
I'm
hoping that by assigning a certain set of keys to the pedals and
reserving
the rest for PD to control that everything can simply work together,
if PD
can output keyboard letters/numbers. Everything would run from the
same
computer. Again, please forgive me if i sound completely ignorant
regarding
what's possible with PD... Many thanks again!
On Tue, Apr 29, 2008 at 6:13 PM, Andrew Turley aturley@acm.org
wrote:
The [fiddle~] and [bonk~] objects should let you capture all of the information that you are looking for. As far as getting that information from Pd to Liquid Media, that may be the tricky part.
Your
current pedals are sending keystrokes straight to Liquid Media. Do
you
know if there is a way to control Liquid Media via MIDI messages, or some sort of message sent via a socket connection (maybe OSC)? If
so,
then you might be able to use Pd to capture your pedal input and
then
combine that with the information from [fiddle~] and/or [bonk~] to send the appropriate information to Liquid Media. If the only way to control Liquid Media is through keystrokes, then you might need to write some software that intercepts the keystrokes from the pedals
and
changes them depending on what is going on in Pd.
andy
On Tue, Apr 29, 2008 at 12:18 PM, Hugh Sung hughsung@gmail.com
wrote:
I'm using a presentation program called "Liquid Media", sort of
like
Powerpoint on steroids. Up to this point i had been using
multiple
foot
pedals assigned to different keystrokes to trigger visual cues,
like
video
clips, animated GIFs and the like, during live classical music
piano
recitals.
I'd like to incorporate the dynamic capabilities of realtime
visual
sync,
with the ability to easily "change the scene", as it were. i'm
hoping
that
someone can help me come up with a combined solution where PD acts
as
an
audio trigger controller, triggering keystroke events based on
pitch
ranges,
dynamic peaks and rhythmic events, and Liquid Media accepts those PD-generated keystrokes to trigger correlating visual cues. Each
scene
would comprise of specific trigger-able events, and the foot
switch
would
allow for a manual "change of scene". i'm hoping to accomplish
two
things
with this combined approach:
- make it a little easier to incorporate visuals synced to audio
events
- allow for easy scene transitions, to better compliment the
"narrative"
structure of classical compositions
Hope this makes sense...thanks all for your input so far! (and
patience!)
On Tue, Apr 29, 2008 at 2:45 PM, Dan Wilcox danomatika@gmail.com
wrote:
> > As for forwarding keypresses and mouse events to an application
...
> > I have done this in Windows to control FLStudio (aka
FruityLoops) by
sending menu events and keypresses through python windows
scripting,
pywinauto, and osc in python. > > I have info and code on my website. It should still work but
it's a
bit
old (2006) and I don't use Windows anymore. > > > > > On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote: > fiddle~ comes with pd (look in the "extra" folder), there is a
help
> patch that will explain how it works. > hope this helps. > marius. > > Hugh Sung wrote: > > i'm not familiar with that program - can you provide a link to
fiddle~,
> > or any other program that can easily do what i'm looking for
with
all
> > the various parameters? > > > > Many thanks! > > > > On Tue, Apr 29, 2008 at 12:50 PM, marius schebella > > <marius.schebella@gmail.com <mailto:marius.schebella@gmail.com
wrote:
> > > > Hugh Sung wrote: > > > > Actually, this isn't for speech recognition - i'm trying to
come
> > up with a simple way to use key triggers to activate visuals. > > > > For example: if the piano plays A440, i'm presuming PD can be > > configured to recognize that pitch with a microphone input and > > then echo out a text string - the letter "A", perhaps. > > > > > > in that case, you best use fiddle~. what's the other program?
does
> > it accept network communication? then you can send the letters
over
> > tcp or udp connection (netsend). > > marius. > > > > > > > > > > -- > > Hugh Sung > > www.hughsung.com http://www.hughsung.com > > www.TabletPCMusician.com http://www.TabletPCMusician.com > > www.musicmeetstech.com http://www.musicmeetstech.com > > > > > > --- > Dan Wilcox > robotcowboy.com
-- Hugh Sung www.hughsung.com
www.TabletPCMusician.com www.musicmeetstech.com _______________________________________________ PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-- Hugh Sung www.hughsung.com www.TabletPCMusician.com www.musicmeetstech.com
-- Hugh Sung www.hughsung.com www.TabletPCMusician.com www.musicmeetstech.com
In real time? Can that be done?
On Tue, May 6, 2008 at 11:35 PM, Andy Farnell padawan12@obiwannabe.co.uk wrote:
Write them to a text file then import them.
On Tue, 6 May 2008 22:54:54 -0400 "Hugh Sung" hughsung@gmail.com wrote:
We're having a neat discussion about this in the Pure Data forum now: http://puredata.hurleur.com/sujet-1686-fiddle-piano
jon2211 came up with a neat patch that demonstrates how well fiddle~ recognizes notes played on a piano within a given range, triggering a
key
text output. Another poster showed an example of modifying that to
trigger
with envelope/volume changes. Of course, we're still stuck with how to
port
the key stroke messages to an external application.
Would someone be willing to write up a python script that can intercept
Pure
Data text echoes and forward them to other concurrently running applications, like PowerPoint or Word? This is the missing link in the puzzle - heck, i'd even be willing to pay someone if they could come up
with
a useful workaround!
Keeping my fingers crossed... Many thanks as always!
On Wed, Apr 30, 2008 at 6:11 AM, Olivier Heinry olivier@heinry.fr
wrote:
Le Tue, 29 Apr 2008 18:50:33 -0400, "Hugh Sung" hughsung@gmail.com a écrit :
Python might bring you the right solution:
http://www.rutherfurd.net/python/sendkeys/
You can either send your letters/keystrokes from inside pd using the
pyext
external that loads your python script
or send messages via OSC to a python script that implements the ixi software simpleOSC lib
http://www.ixi-software.net/content/backyard.html
++ O.
Here's one scenario i'd like to see:
- Pedal assigned [PgDn] keystroke within Liquid Media, which
triggers
scene change 2. PD outputs letters A-Z, depending on either volume or pitch
frequency
ranges input from audio source (microphone into acoustic piano) 3. A given scene within Liquid Media would then interpret the
PD-generated
keystrokes and trigger image events based on letter key "depressed"
by
PD.
You'd have the ease of image generation (a la Powerpoint) with the
speed
and
power of PD to map the sound of acoustic music. Just my newbie
imagination
going here.
On Tue, Apr 29, 2008 at 6:46 PM, Hugh Sung hughsung@gmail.com
wrote:
The pedals basically operate as programmable usb keyboard
extensions,
where you can custom assign any keystroke to any pedal (up to 12).
I'm
hoping that by assigning a certain set of keys to the pedals and
reserving
the rest for PD to control that everything can simply work
together,
if PD
can output keyboard letters/numbers. Everything would run from
the
same
computer. Again, please forgive me if i sound completely ignorant
regarding
what's possible with PD... Many thanks again!
On Tue, Apr 29, 2008 at 6:13 PM, Andrew Turley aturley@acm.org
wrote:
The [fiddle~] and [bonk~] objects should let you capture all of
the
information that you are looking for. As far as getting that information from Pd to Liquid Media, that may be the tricky
part.
Your
current pedals are sending keystrokes straight to Liquid Media.
Do
you
know if there is a way to control Liquid Media via MIDI
messages, or
some sort of message sent via a socket connection (maybe OSC)?
If
so,
then you might be able to use Pd to capture your pedal input and
then
combine that with the information from [fiddle~] and/or [bonk~]
to
send the appropriate information to Liquid Media. If the only
way to
control Liquid Media is through keystrokes, then you might need
to
write some software that intercepts the keystrokes from the
pedals
and
changes them depending on what is going on in Pd.
andy
On Tue, Apr 29, 2008 at 12:18 PM, Hugh Sung hughsung@gmail.com
wrote:
> I'm using a presentation program called "Liquid Media", sort
of
like
> Powerpoint on steroids. Up to this point i had been using
multiple
foot > pedals assigned to different keystrokes to trigger visual
cues,
like
video > clips, animated GIFs and the like, during live classical music
piano
> recitals. > > I'd like to incorporate the dynamic capabilities of realtime
visual
sync, > with the ability to easily "change the scene", as it were.
i'm
hoping
that > someone can help me come up with a combined solution where PD
acts
as
an > audio trigger controller, triggering keystroke events based on
pitch
ranges, > dynamic peaks and rhythmic events, and Liquid Media accepts
those
> PD-generated keystrokes to trigger correlating visual cues.
Each
scene > would comprise of specific trigger-able events, and the foot
switch
would > allow for a manual "change of scene". i'm hoping to
accomplish
two
things > with this combined approach: > > 1. make it a little easier to incorporate visuals synced to
audio
events > 2. allow for easy scene transitions, to better compliment the "narrative" > structure of classical compositions > > Hope this makes sense...thanks all for your input so far! (and patience!) > > > > On Tue, Apr 29, 2008 at 2:45 PM, Dan Wilcox <
danomatika@gmail.com>
wrote: > > > > As for forwarding keypresses and mouse events to an
application
...
> > > > I have done this in Windows to control FLStudio (aka
FruityLoops) by
> sending menu events and keypresses through python windows
scripting,
> pywinauto, and osc in python. > > > > I have info and code on my website. It should still work
but
it's a
bit > old (2006) and I don't use Windows anymore. > > > > > > > > > > On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote: > > fiddle~ comes with pd (look in the "extra" folder), there is
a
help
> > patch that will explain how it works. > > hope this helps. > > marius. > > > > Hugh Sung wrote: > > > i'm not familiar with that program - can you provide a
link to
fiddle~, > > > or any other program that can easily do what i'm looking
for
with
all > > > the various parameters? > > > > > > Many thanks! > > > > > > On Tue, Apr 29, 2008 at 12:50 PM, marius schebella > > > <marius.schebella@gmail.com <mailto:
marius.schebella@gmail.com
wrote: > > > > > > Hugh Sung wrote: > > > > > > Actually, this isn't for speech recognition - i'm trying
to
come
> > > up with a simple way to use key triggers to activate
visuals.
> > > > > > For example: if the piano plays A440, i'm presuming PD can
be
> > > configured to recognize that pitch with a microphone input
and
> > > then echo out a text string - the letter "A", perhaps. > > > > > > > > > in that case, you best use fiddle~. what's the other
program?
does
> > > it accept network communication? then you can send the
letters
over > > > tcp or udp connection (netsend). > > > marius. > > > > > > > > > > > > > > > -- > > > Hugh Sung > > > www.hughsung.com http://www.hughsung.com > > > www.TabletPCMusician.com http://www.TabletPCMusician.com > > > www.musicmeetstech.com http://www.musicmeetstech.com > > > > > > > > > > > > --- > > Dan Wilcox > > robotcowboy.com > > > > -- > Hugh Sung > www.hughsung.com > > www.TabletPCMusician.com > www.musicmeetstech.com > _______________________________________________ > PD-list@iem.at mailing list > UNSUBSCRIBE and account-management -> > http://lists.puredata.info/listinfo/pd-list > >
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-- Hugh Sung www.hughsung.com www.TabletPCMusician.com www.musicmeetstech.com
-- Hugh Sung www.hughsung.com www.TabletPCMusician.com www.musicmeetstech.com
-- Use the source
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Dude I sent you this info already ... its called pywinauto, a Python automation suite for controlling Windows apps. You can send key and mouse events such as opening a file dialog, "typing in the filename", and hitting enter. There are plenty of straightforward info in the documentation.
I paired that with the aforementioned ixi software's simpleOSC library. There are examples around.
Once again. if you want a specific example, download the zip on this page of my website where I use PD to send and receive control signals to Python/pywinauto over OSC to control FLStudio.
I know you a "newb" and all, but that dosen't mean you can't get your hands dirty and learn a bit. When I first put that all together I did not understand OSC and had never used Python or automation software before.
On Tue, 2008-05-06 at 22:54 -0400, Hugh Sung wrote:
We're having a neat discussion about this in the Pure Data forum now: http://puredata.hurleur.com/sujet-1686-fiddle-piano
jon2211 came up with a neat patch that demonstrates how well fiddle~ recognizes notes played on a piano within a given range, triggering a key text output. Another poster showed an example of modifying that to trigger with envelope/volume changes. Of course, we're still stuck with how to port the key stroke messages to an external application.
Would someone be willing to write up a python script that can intercept Pure Data text echoes and forward them to other concurrently running applications, like PowerPoint or Word? This is the missing link in the puzzle - heck, i'd even be willing to pay someone if they could come up with a useful workaround!
Keeping my fingers crossed... Many thanks as always!
On Wed, Apr 30, 2008 at 6:11 AM, Olivier Heinry olivier@heinry.fr wrote:
Le Tue, 29 Apr 2008 18:50:33 -0400, "Hugh Sung" <hughsung@gmail.com> a écrit : Python might bring you the right solution: http://www.rutherfurd.net/python/sendkeys/ You can either send your letters/keystrokes from inside pd using the pyext external that loads your python script or send messages via OSC to a python script that implements the ixi software simpleOSC lib http://www.ixi-software.net/content/backyard.html ++ O. > Here's one scenario i'd like to see: > > 1. Pedal assigned [PgDn] keystroke within Liquid Media, which triggers > scene change > 2. PD outputs letters A-Z, depending on either volume or pitch frequency > ranges input from audio source (microphone into acoustic piano) > 3. A given scene within Liquid Media would then interpret the PD-generated > keystrokes and trigger image events based on letter key "depressed" by PD. > > You'd have the ease of image generation (a la Powerpoint) with the speed and > power of PD to map the sound of acoustic music. Just my newbie imagination > going here. > > On Tue, Apr 29, 2008 at 6:46 PM, Hugh Sung <hughsung@gmail.com> wrote: > > > The pedals basically operate as programmable usb keyboard extensions, > > where you can custom assign any keystroke to any pedal (up to 12). I'm > > hoping that by assigning a certain set of keys to the pedals and reserving > > the rest for PD to control that everything can simply work together, if PD > > can output keyboard letters/numbers. Everything would run from the same > > computer. Again, please forgive me if i sound completely ignorant regarding > > what's possible with PD... > > Many thanks again! > > > > > > On Tue, Apr 29, 2008 at 6:13 PM, Andrew Turley <aturley@acm.org> wrote: > > > > > The [fiddle~] and [bonk~] objects should let you capture all of the > > > information that you are looking for. As far as getting that > > > information from Pd to Liquid Media, that may be the tricky part. Your > > > current pedals are sending keystrokes straight to Liquid Media. Do you > > > know if there is a way to control Liquid Media via MIDI messages, or > > > some sort of message sent via a socket connection (maybe OSC)? If so, > > > then you might be able to use Pd to capture your pedal input and then > > > combine that with the information from [fiddle~] and/or [bonk~] to > > > send the appropriate information to Liquid Media. If the only way to > > > control Liquid Media is through keystrokes, then you might need to > > > write some software that intercepts the keystrokes from the pedals and > > > changes them depending on what is going on in Pd. > > > > > > andy > > > > > > On Tue, Apr 29, 2008 at 12:18 PM, Hugh Sung <hughsung@gmail.com> wrote: > > > > I'm using a presentation program called "Liquid Media", sort of like > > > > Powerpoint on steroids. Up to this point i had been using multiple > > > foot > > > > pedals assigned to different keystrokes to trigger visual cues, like > > > video > > > > clips, animated GIFs and the like, during live classical music piano > > > > recitals. > > > > > > > > I'd like to incorporate the dynamic capabilities of realtime visual > > > sync, > > > > with the ability to easily "change the scene", as it were. i'm hoping > > > that > > > > someone can help me come up with a combined solution where PD acts as > > > an > > > > audio trigger controller, triggering keystroke events based on pitch > > > ranges, > > > > dynamic peaks and rhythmic events, and Liquid Media accepts those > > > > PD-generated keystrokes to trigger correlating visual cues. Each > > > scene > > > > would comprise of specific trigger-able events, and the foot switch > > > would > > > > allow for a manual "change of scene". i'm hoping to accomplish two > > > things > > > > with this combined approach: > > > > > > > > 1. make it a little easier to incorporate visuals synced to audio > > > events > > > > 2. allow for easy scene transitions, to better compliment the > > > "narrative" > > > > structure of classical compositions > > > > > > > > Hope this makes sense...thanks all for your input so far! (and > > > patience!) > > > > > > > > > > > > > > > > On Tue, Apr 29, 2008 at 2:45 PM, Dan Wilcox <danomatika@gmail.com> > > > wrote: > > > > > > > > > > As for forwarding keypresses and mouse events to an application ... > > > > > > > > > > I have done this in Windows to control FLStudio (aka FruityLoops) by > > > > sending menu events and keypresses through python windows scripting, > > > > pywinauto, and osc in python. > > > > > > > > > > I have info and code on my website. It should still work but it's a > > > bit > > > > old (2006) and I don't use Windows anymore. > > > > > > > > > > > > > > > > > > > > > > > > > On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote: > > > > > fiddle~ comes with pd (look in the "extra" folder), there is a help > > > > > patch that will explain how it works. > > > > > hope this helps. > > > > > marius. > > > > > > > > > > Hugh Sung wrote: > > > > > > i'm not familiar with that program - can you provide a link to > > > fiddle~, > > > > > > or any other program that can easily do what i'm looking for with > > > all > > > > > > the various parameters? > > > > > > > > > > > > Many thanks! > > > > > > > > > > > > On Tue, Apr 29, 2008 at 12:50 PM, marius schebella > > > > > > <marius.schebella@gmail.com <mailto:marius.schebella@gmail.com>> > > > wrote: > > > > > > > > > > > > Hugh Sung wrote: > > > > > > > > > > > > Actually, this isn't for speech recognition - i'm trying to come > > > > > > up with a simple way to use key triggers to activate visuals. > > > > > > > > > > > > For example: if the piano plays A440, i'm presuming PD can be > > > > > > configured to recognize that pitch with a microphone input and > > > > > > then echo out a text string - the letter "A", perhaps. > > > > > > > > > > > > > > > > > > in that case, you best use fiddle~. what's the other program? does > > > > > > it accept network communication? then you can send the letters > > > over > > > > > > tcp or udp connection (netsend). > > > > > > marius. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > > Hugh Sung > > > > > > www.hughsung.com <http://www.hughsung.com> > > > > > > www.TabletPCMusician.com <http://www.TabletPCMusician.com> > > > > > > www.musicmeetstech.com <http://www.musicmeetstech.com> > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > --- > > > > > Dan Wilcox > > > > > robotcowboy.com > > > > > > > > > > > > > > > > -- > > > > Hugh Sung > > > > www.hughsung.com > > > > > > > > www.TabletPCMusician.com > > > > www.musicmeetstech.com > > > > _______________________________________________ > > > > PD-list@iem.at mailing list > > > > UNSUBSCRIBE and account-management -> > > > > http://lists.puredata.info/listinfo/pd-list > > > > > > > > > > > > > > _______________________________________________ > > > PD-list@iem.at mailing list > > > UNSUBSCRIBE and account-management -> > > > http://lists.puredata.info/listinfo/pd-list > > > > > > > > > > > -- > > Hugh Sung > > www.hughsung.com > > www.TabletPCMusician.com > > www.musicmeetstech.com > > > > >-- Hugh Sung www.hughsung.com www.TabletPCMusician.com www.musicmeetstech.com
Dan Wilcox robotcowboy.com
Hi I've developped this external for windows last year, this emulate keystrokes, the result of ascii code depends on your hardware, it uses kbd_event API, but I believe the python script from previous messages that uses sendkey would be much better.
You will have to use messages like this:
[downkey 18, key 9, key 9, upkey 18( for alt+tab combination.
Use it at your own risk, good luck.
Hugh Sung a écrit :
Here's one scenario i'd like to see:
- Pedal assigned [PgDn] keystroke within Liquid Media, which triggers
scene change 2. PD outputs letters A-Z, depending on either volume or pitch frequency ranges input from audio source (microphone into acoustic piano) 3. A given scene within Liquid Media would then interpret the PD-generated keystrokes and trigger image events based on letter key "depressed" by PD.
You'd have the ease of image generation (a la Powerpoint) with the speed and power of PD to map the sound of acoustic music. Just my newbie imagination going here.
On Tue, Apr 29, 2008 at 6:46 PM, Hugh Sung <hughsung@gmail.com mailto:hughsung@gmail.com> wrote:
The pedals basically operate as programmable usb keyboard extensions, where you can custom assign any keystroke to any pedal (up to 12). I'm hoping that by assigning a certain set of keys to the pedals and reserving the rest for PD to control that everything can simply work together, if PD can output keyboard letters/numbers. Everything would run from the same computer. Again, please forgive me if i sound completely ignorant regarding what's possible with PD... Many thanks again! On Tue, Apr 29, 2008 at 6:13 PM, Andrew Turley <aturley@acm.org <mailto:aturley@acm.org>> wrote: The [fiddle~] and [bonk~] objects should let you capture all of the information that you are looking for. As far as getting that information from Pd to Liquid Media, that may be the tricky part. Your current pedals are sending keystrokes straight to Liquid Media. Do you know if there is a way to control Liquid Media via MIDI messages, or some sort of message sent via a socket connection (maybe OSC)? If so, then you might be able to use Pd to capture your pedal input and then combine that with the information from [fiddle~] and/or [bonk~] to send the appropriate information to Liquid Media. If the only way to control Liquid Media is through keystrokes, then you might need to write some software that intercepts the keystrokes from the pedals and changes them depending on what is going on in Pd. andy On Tue, Apr 29, 2008 at 12:18 PM, Hugh Sung <hughsung@gmail.com <mailto:hughsung@gmail.com>> wrote: > I'm using a presentation program called "Liquid Media", sort of like > Powerpoint on steroids. Up to this point i had been using multiple foot > pedals assigned to different keystrokes to trigger visual cues, like video > clips, animated GIFs and the like, during live classical music piano > recitals. > > I'd like to incorporate the dynamic capabilities of realtime visual sync, > with the ability to easily "change the scene", as it were. i'm hoping that > someone can help me come up with a combined solution where PD acts as an > audio trigger controller, triggering keystroke events based on pitch ranges, > dynamic peaks and rhythmic events, and Liquid Media accepts those > PD-generated keystrokes to trigger correlating visual cues. Each scene > would comprise of specific trigger-able events, and the foot switch would > allow for a manual "change of scene". i'm hoping to accomplish two things > with this combined approach: > > 1. make it a little easier to incorporate visuals synced to audio events > 2. allow for easy scene transitions, to better compliment the "narrative" > structure of classical compositions > > Hope this makes sense...thanks all for your input so far! (and patience!) > > > > On Tue, Apr 29, 2008 at 2:45 PM, Dan Wilcox <danomatika@gmail.com <mailto:danomatika@gmail.com>> wrote: > > > > As for forwarding keypresses and mouse events to an application ... > > > > I have done this in Windows to control FLStudio (aka FruityLoops) by > sending menu events and keypresses through python windows scripting, > pywinauto, and osc in python. > > > > I have info and code on my website. It should still work but it's a bit > old (2006) and I don't use Windows anymore. > > > > > > > > > > On Tue, 2008-04-29 at 13:05 -0400, marius schebella wrote: > > fiddle~ comes with pd (look in the "extra" folder), there is a help > > patch that will explain how it works. > > hope this helps. > > marius. > > > > Hugh Sung wrote: > > > i'm not familiar with that program - can you provide a link to fiddle~, > > > or any other program that can easily do what i'm looking for with all > > > the various parameters? > > > > > > Many thanks! > > > > > > On Tue, Apr 29, 2008 at 12:50 PM, marius schebella > > > <marius.schebella@gmail.com <mailto:marius.schebella@gmail.com> <mailto:marius.schebella@gmail.com <mailto:marius.schebella@gmail.com>>> wrote: > > > > > > Hugh Sung wrote: > > > > > > Actually, this isn't for speech recognition - i'm trying to come > > > up with a simple way to use key triggers to activate visuals. > > > > > > For example: if the piano plays A440, i'm presuming PD can be > > > configured to recognize that pitch with a microphone input and > > > then echo out a text string - the letter "A", perhaps. > > > > > > > > > in that case, you best use fiddle~. what's the other program? does > > > it accept network communication? then you can send the letters over > > > tcp or udp connection (netsend). > > > marius. > > > > > > > > > > > > > > > -- > > > Hugh Sung > > > www.hughsung.com <http://www.hughsung.com> <http://www.hughsung.com> > > > www.TabletPCMusician.com <http://www.TabletPCMusician.com> <http://www.TabletPCMusician.com> > > > www.musicmeetstech.com <http://www.musicmeetstech.com> <http://www.musicmeetstech.com> > > > > > > > > > > > > --- > > Dan Wilcox > > robotcowboy.com <http://robotcowboy.com> > > > > -- > Hugh Sung > www.hughsung.com <http://www.hughsung.com> > > www.TabletPCMusician.com <http://www.TabletPCMusician.com> > www.musicmeetstech.com <http://www.musicmeetstech.com> > _______________________________________________ > PD-list@iem.at <mailto:PD-list@iem.at> mailing list > UNSUBSCRIBE and account-management -> > http://lists.puredata.info/listinfo/pd-list > > _______________________________________________ PD-list@iem.at <mailto:PD-list@iem.at> mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list -- Hugh Sung www.hughsung.com <http://www.hughsung.com> www.TabletPCMusician.com <http://www.TabletPCMusician.com> www.musicmeetstech.com <http://www.musicmeetstech.com>-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Patrice Colet a écrit :
Hi I've developped this external for windows last year, this emulate keystrokes, the result of ascii code depends on your hardware, it uses kbd_event API, but I believe the python script from previous messages that uses sendkey would be much better.
For example ou will have to use messages like this:
[downkey 18, key 9, key 9, upkey 18( to emulate alt+tab+tab combination, .
Use it at your own risk, good luck.
ooops, I wanted to say "the result of the string sent to the object depends on the hardware"
and not "the result of ascii code depends on your hardware".
also if you click on the message that emulate keystroke, the focused window will still be the pd patch because it focuses on mouse release, also the object doesn't seem to work if we use [key] for triggering the message, so I put a delay between the bang and the message, then the message is sent after the mouse click is released.
an example is attached
thankyou
Ok...please bear with me, as i continue to slowly get familiar with learning PD! I've managed to get a working start, using a sendkey VB script graciously written up by someone from the PD forum. Using bonk~ and moses~, i'm able to split incoming audio to loud and soft levels, which then send commands to open two separate versions of the sendkey vb script - one to type a key for loud, the other to type something different for the soft sound input. Problem is that the script seems a bit slow - there's a noticeable delay when sending signals from PD to the VB script then to the final program receiving the text strings. i'm hoping to get speedier results either going the sendOSC route, or by using one of Patrice Colet's external scripts. i'm including a link to the zipped copy of my PD project from the PD forum: http://puredata.hurleur.com/viewtopic.php?pid=7514#p7514
So here is my new round of still-newbie questions:
- How do i install and use the external scripts winkey and/or kbdstroke?
Or do i need to just call those externals from within pd somehow? 2. I'm trying to emulate Dan Wilcox's (aka: robotcowboy's) sendOSC method- i seem to be able to connect to the OSC port, but i don't know how to activate the 'sendkeys' script. I get a 'no method for 'bang'' error when i try to test with a straight message bang:
sendOSC: connected to port localhost:9001 (hSock=11213640) protocol = UDP
error: sendOSC: no method for 'bang'
Can someone hold my hand and show me how to call 'sendkeys' from within PD to send strings/text to another application? i'm hoping this method will be faster than the VB script method...
i'm really sorry for hogging up the PD-list with this - i'm getting so close now! Thanks ever so much for all your collective help!
Hugh
On Thu, May 8, 2008 at 6:02 PM, Patrice Colet pat@mamalala.org wrote:
Patrice Colet a écrit :
Hi I've developped this external for windows last year, this emulate keystrokes, the result of ascii code depends on your hardware, it uses kbd_event API, but I believe the python script from previous messages that uses sendkey would be much better.
For example ou will have to use messages like this:
[downkey 18, key 9, key 9, upkey 18( to emulate alt+tab+tab combination, .
Use it at your own risk, good luck.
ooops, I wanted to say "the result of the string sent to the object depends on the hardware"
and not "the result of ascii code depends on your hardware".
also if you click on the message that emulate keystroke, the focused window will still be the pd patch because it focuses on mouse release, also the object doesn't seem to work if we use [key] for triggering the message, so I put a delay between the bang and the message, then the message is sent after the mouse click is released.
an example is attached
thankyou
Hugh Sung a écrit :
- How do i install and use the external scripts winkey and/or
kbdstroke? Or do i need to just call those externals from within pd somehow?
sorry, I forgot to mention how to install it, put kbdstroke.dll in extra folder, and kbdstroke-help.pd in doc/5.reference, reload pd, it installed.
This external is relatively fast, I didn't notice any particular latency.
Hi,
"sigmund~" and "fiddle~" are objects to analyse the microphone signal to find the fundamental frequency.
You can use the "moses" and "select" objects to make something happend when the frequency is into a particular range.
I do not know how to simulate a keyboard input.
Cheers, Damien.
Hugh Sung a écrit :
Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals.
For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps.
I have a presentation program that can be configured to accept keyboard strokes to activate visual events.
Does that example make sense? PD sounds like it should be able to handle that capability pretty easily...
On Tue, Apr 29, 2008 at 12:14 PM, marius schebella <marius.schebella@gmail.com mailto:marius.schebella@gmail.com> wrote:
speech to text in opensource software is still poor, and even poorer for languages other than english. afaik pd also doesnot have a builtin solution for that. you could try to use a different program for that and parse the text, or you could try more basic approaches within pd (like pitch recognition, amplitude/envelope following, or rhythm/melody recognition. I am sure there are some commercial solutions out there (like in the iphone) that have good speech-to-text recognition, including dictionaries. sorry that this probably is no real help for you, maybe someone else knows more? marius. Hugh Sung wrote: Hello! I'm new to PD and was wondering if i could get a "working start" - i'd like to someday be able to create a PD applet that can take audio input from a microphone and convert the sounds to text that can simultaneously be applied to a concurrently running program, like a word processor or a presentation program that accepts alpha-numeric input to trigger events. Can someone create a PD example that does the following: 1. echo Alpha/numeric characters based on variable pitch 2. echo alpha/numeric characters based on volume 3. echo alpha/numeric characters based on rhythm (ie, number of pulses per second or minute) Sorry if this sounds simplistic - if i can see an example of a PD patch that can incorporate these elements, that will go a LONG way to helping me learn how to program PD on my own! Thanks in advance for your help and suggestions, PD list! -- Hugh Sung www.hughsung.com <http://www.hughsung.com> <http://www.hughsung.com> www.TabletPCMusician.com <http://www.TabletPCMusician.com> <http://www.TabletPCMusician.com> www.musicmeetstech.com <http://www.musicmeetstech.com> <http://www.musicmeetstech.com> ------------------------------------------------------------------------ _______________________________________________ PD-list@iem.at <mailto:PD-list@iem.at> mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
What are you using for the "presentation" software?
You can look at bonk~ and fiddle~ to process sounds, but if your just showing images or video you could do that part in PD also using Gem or PDP.
.b.
Hugh Sung wrote:
Actually, this isn't for speech recognition - i'm trying to come up with a simple way to use key triggers to activate visuals.
For example: if the piano plays A440, i'm presuming PD can be configured to recognize that pitch with a microphone input and then echo out a text string - the letter "A", perhaps.
I have a presentation program that can be configured to accept keyboard strokes to activate visual events.
Does that example make sense? PD sounds like it should be able to handle that capability pretty easily...
On Tue, Apr 29, 2008 at 12:14 PM, marius schebella <marius.schebella@gmail.com mailto:marius.schebella@gmail.com> wrote:
speech to text in opensource software is still poor, and even poorer for languages other than english. afaik pd also doesnot have a builtin solution for that. you could try to use a different program for that and parse the text, or you could try more basic approaches within pd (like pitch recognition, amplitude/envelope following, or rhythm/melody recognition. I am sure there are some commercial solutions out there (like in the iphone) that have good speech-to-text recognition, including dictionaries. sorry that this probably is no real help for you, maybe someone else knows more? marius. Hugh Sung wrote: Hello! I'm new to PD and was wondering if i could get a "working start" - i'd like to someday be able to create a PD applet that can take audio input from a microphone and convert the sounds to text that can simultaneously be applied to a concurrently running program, like a word processor or a presentation program that accepts alpha-numeric input to trigger events. Can someone create a PD example that does the following: 1. echo Alpha/numeric characters based on variable pitch 2. echo alpha/numeric characters based on volume 3. echo alpha/numeric characters based on rhythm (ie, number of pulses per second or minute) Sorry if this sounds simplistic - if i can see an example of a PD patch that can incorporate these elements, that will go a LONG way to helping me learn how to program PD on my own! Thanks in advance for your help and suggestions, PD list! -- Hugh Sung www.hughsung.com <http://www.hughsung.com> <http://www.hughsung.com> www.TabletPCMusician.com <http://www.TabletPCMusician.com> <http://www.TabletPCMusician.com> www.musicmeetstech.com <http://www.musicmeetstech.com> <http://www.musicmeetstech.com> ------------------------------------------------------------------------ _______________________________________________ PD-list@iem.at <mailto:PD-list@iem.at> mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list-- Hugh Sung www.hughsung.com http://www.hughsung.com www.TabletPCMusician.com http://www.TabletPCMusician.com www.musicmeetstech.com http://www.musicmeetstech.com
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-
Andrew Turley -
Andy Farnell -
B. Bogart -
Damien Henry - Voxler -
Dan Wilcox -
Hugh Sung -
marius schebella -
Olivier Heinry -
Patrice Colet