Player has several audio related interfaces:
-speech: implemented in the speech drivers that takes care of launching and
killing festival along with the proper communication with it.
-speech_recognition: The implementation depends on sphinx2
-audio: Currently seems disabled
The fact is that those drivers are managed in a very model independent way by
Player. So, I don't see any need to take those drivers and implement them in
gazebo (say copy and paste).
What I meant is that through Player it is perfectly possible to integrate text
to speech and sound to a simulation. For instance, the wheels turns in Gazebo
model and the client knows it thought the gazebo driver, then use the audio
driver to play some funny sound.
The only problem is that Gazebo has 2 interfaces: the one with player and the
one named libgazebo.
The second interface needs audio?
In fact, we need that second interface?
Why is Gazebo project maintaining libgazebo instead of modify wxgazebo to use
player (similar to playerv)?
I think that most people use Gazebo because it is used "the official" 3d sim
for player, why don't accept this fact and stop maintaining libgazebo ?
> In fact, we need that second interface?
> Why is Gazebo project maintaining libgazebo instead of modify wxgazebo
> to use player (similar to playerv)?
> I think that most people use Gazebo because it is used "the official" 3d
> sim for player, why don't accept this fact and stop maintaining
> libgazebo ?
I do use libgazebo because player does not fit my needs (in fact, I already
have a robot control layer, no need for player thankyouverymuch).
Moreover, if you don't use player, it is easier to extend a simple library
like libgazebo instead of having to adapt both player *and* gazebo source