A few weeks ago I bought a Nabaztag and I was impressed by functionalities and possibilities of the “first second connected rabbit“.
I started playing with it and after a few days I decided to explore a little bit deeper in the services that I could run.
I came across opennab that is basically a proxy server for Nabaztags and offers a smart enough plugin architecture to develop custom services.
Since I am a big fan of Jaiku the first thing I was thinking was: Since I always have a web page open on my Jaiku stream why cannot the Nabaztag check that stream on my behalf and read them to me using some kind of TTS?
This is how NabazJaiku was born.
NabazJaiku is a (very) quick and (very) dirty php script that plugs in the opennab architecture and monitors my Jaiku streams. When updates are available they are formatted in a plain text file that is processed by a TTS engine and then sent to the Nabaztag.
If some of you is interested in having a little bit more technical details on how the thing works just drop a comment. I will eventually publish a dedicated page when I will have finished some fine tuning on the TTS speech engine.
Yesterday night I put the whole thing on my server and since then I am listening to Jaiku updates through my Nabaztag.
The interesting thing here is that there is an interaction between a web service (Jaiku) and a real device (Nabaztag).
I think there are plenty of applications like this for a Nabaztag or other connected devices.
A video of an early version of the application is embedded in this post. The version I was using did not have any control of the TTS engine through the SSML (Speech Synthesis Markup Language) so you will hear a stream of voice with no pauses between Jaiku posts. The last version of the script has much more control.
One of the issue that need to be solved is trying to identify in which language Jaiku posts are written.. This is goin to be tought and will probably never happen.Why do I blog this? You know, at the end of the day I’m a geek.