On 03/20/2012 04:29 PM, Romain Beauxis wrote:
Hi Andre!
2012/3/20 andre castro<[email protected]>:
Since, I have been asking a questions to liquidsoap's list for this
project, I would also like you folks to know about it.
For the last couple of months I have slowly been developing a trimester
project whereby the wealth and diversity of audio material present in
Archive.org is explored in an unusual way, by running an online radio
station based on such material. Using (of course) liquidsoap!
The project is very betaish. It's name, visual identity, and loads of
details are not yet defined, and you might experience some hiccups and
bugs in the stream, but it is moving.
The stream url is http://pzwart1.wdka.hro.nl:8000/radio_blind_archive.
You can also listen to it in the browser(at least for the next couple of
days) http://pzwart3.wdka.hro.nl/~acastro/blind-archive/andre.html
I would like very much to hear what you people think about. And if you
have suggestions, critiques, possible contributions, i would be
delighted to hear them :)
I think it is a very interesting project! I'd be very interested to
hear about how you compose the stream and pick tracks.
Yes, that will be included when the project is more advanced. But to
give a brief idea:
* it tries to emulate a real radio station, run by humans (will more
developed with jingles, hours announcements,etc)
* in this radio "we' have shows. Each show has a genre-topic. Currently
'we' are trying a few programs: news, cooking-old records,
science-ambient, poetry-hiphop, weather forecast-classical music.
* programs come in pairs of topic-music genre due to the mechanism
according to which station works.
* for each program a specific RSS content is gathered daily. This
content (its words) become search terms on archive.org.
Taking the example of the news program:
1)Feeds published by the NYTimes are gathered.
2)For word a search is made on archive.org for soundfiles tagged under
that word AND under spoken word genre.
The result is (mostly) is a sequence of files with similar sonic
universe and sometimes referring to some current themes (Iraq, Obama, etc).
On cooking:
text source: stockexchange cooking discussion page;
search criteria: soundfiles from the 78rpm records collection;
So we can say that each program from the stream is built from:
[thematic text source(news feeds, weather forecast feeds, thematic
mailing-lists]
+
[search criteria(old records,hiphop,spoken)]
=
Sound files (from a particular universe)
Yet a bit a convoluted way to explain it but here it is. :)
The reason is that part of the ideas I'd like to explore one day is a
way to gather sort-of web semantics information about audio files and
create streams based on that.
Sounds interesting. Why not going for it ?!?
BTW, one of the original project that liquidsoap was built on was a
net radio picking up random network shares on a microsoft/samba
network on a university campus. David and Sam could talk about more
but I think they has a lot of fun with this project!
That would be very interesting to hear about.
Thank you much for your interest
best
a
Romain
--
_
(` ).
http://www.andrecastro.info/
_( '`.
.=(`( . ) .--
(( (..__.:'-' .+( )
`( ) ( . )
` __.:' `-' '-
--..,___.--,--'-,---..-.--+--.,,-,,...
------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here
http://p.sf.net/sfu/sfd2d-msazure
_______________________________________________
Savonet-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/savonet-users
