[ros-users] Connecting ROS to a software (music) synthesiser?

Raymond Sheh Raymond.Sheh at curtin.edu.au
Tue Nov 11 14:11:07 UTC 2014


On 11/11/2014 8:17 PM, Federico Ferri wrote:
> On 11 Nov 2014 at 07:42:53, Raymond Sheh (raymond.sheh at curtin.edu.au) wrote:
>
>> We're trying to figure out a way to have our robot's beeps be a little
>> more musical and on-the-fly generated so I'm looking for a way of piping
>> ROS messages into a software music synthesiser of some sort (or
>> synthesizer for those of you in the US).
>>
>> I've found lots of stuff about getting ROS to play sounds or to say text
>> via a text-to-speech synth but I'm drawing a bit of a blank on something
>> that is more musical.
>>
>> Before I go to the trouble of connecting ROS to something like
>> Supercollider or Fluidsynth, I was just wondering if anyone here knew of
>> work already being done in this direction. My Google-fu is kinda failing
>> me on this one.
>
> There exists MIDI, which is a standard, consisting of a digital hardware interface and a transmission protocol. It exists since 80s and with this you can interface (almost) any hardware and software synthesizer and tone generator.
>
> There is also the relatively new OSC, which is a protocol with similar purposes to MIDI (perhaps extending MIDI), but using modern networking medium.
>
> I think it can make sense to develop a bridge between ROS and any of those standards.
>
> My two cents.
>
> Cheers,
> Federico Ferri
Hi Federico,

Thanks for the suggestion!

I certainly agree that going the MIDI route would be the "proper" way of 
doing this if we were dealing with hardware synths. I did dabble in some 
MIDI hardware in the past, based on that experience I think that to do 
it properly would be more of an undertaking than I can justify right now 
- especially as there's no MIDI hardware in the loop at this point (nor 
is there intending to be at any point in the future).

I was thinking about the OSC route - both ROS and Supercollider have 
Python interfaces and the Python interface for Supercollider runs 
through OSC. Then just as I ran into ROSOSC, Michael replied to my post. 
:-)

Looks like I have a few options now.

Cheers!

- Raymond

-- 

Dr Raymond Sheh
Senior Lecturer

Curtin University of Technology
Department of Computing
Level 3, Building 314, Kent St
Bentley, Western Australia

+61 8 9266 4269
GPO Box U1987 Perth WA 6845 AUSTRALIA



More information about the ros-users mailing list