Exciting! Late-breaking results
Matthias did a great job. Marsyas is used for MFCC extraction. Rapidminer does a good job on k-means clustering. The results go as colored labels into .kml. GoogleEarth. Thats it!
Interesting to see that the clusters represent certain segments and places of the route being not disturbed by outliers. We have to look into it furthermore … but we are a little bit excited.
Student worker! Rapidminer …
I am happy. I found a very good student who will be using Rapidminer to perform an unsupervised clustering of the audio streams. Hopefully we will find the perfect combination of features and settings. At least the visualization within the GoogleEarth package should be nice to navigate. 10 weeks to go …
Science Pop or Dissemination?
I am giving some interviews these days and I force myself to publish more for everyday people. Why? It seems to be necessary to explain what we scientists do and why some of our results matter beyond pure technological achievements. The omnipresence and re-shaping of media is something which squeezes my mind 24/7. I am running a more general blog about these topics for the Competence Center Computational Culture at DFKI. Unfortunately I do not post frequently because the core science needs time, too. And even worse I have to translate a lot of stuff back and forth between German and English. Acting local is important and allows for face-2-face interaction. But since my main interests may be a little bit in a niche I have to act most of the time on an international scale. Sounds pretty simple but it gives me from time to time some headaches.
Anyway I am doing fine and looking forward for discussion with you guys … see you soon here and there.
Silence
I am reading R. Murray Schafer’s SOUNDSCAPE. A classic book, published first in 1977, still an excellent read. In parallel I stumbled over an advertisement in one of those journals for uebergeek-tekkies. The Sony noise cancellation headphones with integrated AI promise some rare luxury within these days. Silence. Artificial Silence. At a first glance a bit pricy, but for the sake of research I am really considering to buy them soon.
Sounds and Noises from Berlin
I am sorry that I still was not able to perform Urban Sync session in Berlin. But the good news is that there has been activities alike since a couple of years.
E.g the Berlincast.
Or the TunedCity project.
Enjoy!
Soundmap of Cologne
My local newspaper contained an article about people working hard on capturing soundscapes worldwide. The article had a focus on German activities. If you are interested in the Soundmap of Cologne you should check the great blog about the project including cc-licensed samples:
In the air – Visualization of urban data about Madrid
I am reading regularly the Infosthetics blog. Excellent source for the latest and best projects, tools, etc for interactive visualization of large-scale data. The posting about the In the Air project indicates that there will be updates about the coding in Processing which may be of interest for Urban Sync.
Looking forward to Lift09!
Thanks to Steffen for the hint! I will join Lift09 in February. An excellent melting pot to spread my ideas about reality tagging /blogging and the future of emotions. Great time for brainstorming, networking with inspiring people. Maybe also some serious SNOW?
Small excerpt of the webpage:
Lift09 (Geneva, 25-27 Feb. 09) will look back to look ahead, exploring topics like change, solidarity, love, or design, during three days of intense networking and inspiration themed around a simple question: “Where did the future go?“
We were told the future would be about mechanization, computerization, 1984-like nightmares or robots. What did and did not happen? What can we learn from the predictions that never materialized to better look at the future?
Lift gathers international entrepreneurs, artists, managers, researchers, investors, CEOs, designers or ethnologist, people who come to be inspired and meet those who make a difference.
The Berlin Sessions – UdK Sound Studies
Urban Sync is still work in progress. For my Berlin Sessions I am trying to find potential partners. Yesterday I had a very productive and friendly meeting with Prof.Schulze, the mastermind of the UdK Sound Studies department. We talked about each and every aspect of my project, their students, the philosophy and potenial of sonic spaces, soundscapes, interdisciplinary projects, fundraising, commuting, modern life, identities, blogs, micro-blogging, reality mining, … and a joint agenda.
We did not talk for hours. Just 45 minutes have been enough since there was so much unspoken agreement about our research and why the journey into sound and space matters.
A small repertoire of urban gestures
I am working on UrbanScrub by using standard software. Why? I like the concept of bodystorming in order to act out possible usage by early investigations based on mockups. At the current stage I used Fission to auto-split audio streams into small grains. Furthermore I imported these grains into the NN-XT sampler of Reason 3. STEIM is offering the versatile mapping tool junXion which allows for very smooth processing of Wiimote input. By nesting logical operations I was able to generate MIDI output to trigger the grains in NN-XT.
Video footage soon to follow here …
