I would like to share a work which was complete about 1 year ago which
for various reasons is online only today: a video story (short film) by
director Matteo Scerfò  for which I created the sound My part was
done 100% in Linux.
Links (see below for a brief explanation and linux-audio tech. details).
Direct link to ogv:
Audio only ogg:
I created the complete sound(track) as the final work for my electronic
music diploma dissertation last year.
As you can hear, this is not your tipical soundtrack :-). The film was
initially conceived as silent, so my task was to give it sound at all
levels. A more detailed explanation is provided in the accompanying
paper  which I'm afraid is in Italian only, however I'll be happy to
provide details if anyone is curious.
All audio was created on Linux using FLOSS software. All audio was also
generated, i.e. there is no sampled sound - with the exception of TTS
which strictly speaking _does_ use samples.
Audio software used.
- Ardour (version 2 - with various LADSPA effects..)
- Xjadeo (jack-transpor aware video player)
- Pure Data (including Granita glanular synth)
- Festival (speech synthesis - with some MBROLA voices)
- hexter DSSI
Ardour + Xjadeo was my montage workbench. Pure Data was used to generate
all sound ecept or TTS done through Festival. I owe much inspiration to
Andy Farnell and his book Designing Sound  for the concepts, ideas
and the philosophy of generating sound. Rosegarden was used to compose
and sequence the 'musical' parts.
Excluding TTS where wave files were generated and then imported into
Ardour all the rest was directly fed and recorded into the DAW through JACK.
Linux-audio-user mailing list