W W elcome to Etherpad!

This pad text is synchronized as you type, so that everyone viewing this page sees the same text. This allows you to collaborate seamlessly on documents!

Get involved with Etherpad at http://etherpad.org
riting with film 23-2-16
Gijs, Hans, Piero, Sam, An, Manetta

http://pad.constantvzw.org/p/writing_with_film/export/txt

Witte de with: http://www.wdw.nl/event/foreword-calla-henkel-max-pitegoff/
-> use this as goal for today's work?

Dear An Mertens,
I hope this email finds you well. 
In the context of the exhibition Para|Fictions at Witte de With (WdW) Center for Contemporary Art in Rotterdam, the
??
WdW Class of '16 –  the collaborative student co-programming team of Witte de With – would  like to invite you to contribute to our first event. The event will be  centered around the contamination and slippages between artistic  and entrepreneurial language. In Para|Fictions : Foreword,  artists Calla Henkel & Max Pitegoff intensify the abstractions of  notions such as community, participation, space and radical future by  focusing on the language as the object and product of  entrepreneurial startups. The event will zero in on the textuality  of this kind of language and will attempt to expose its mechanisms of  creating and constructing a collective lived fantasy. 
We  invite you to reflect on the above themes, and also, in the spirit of  the Algolit workgroup, we would like you to present your project and  attempt to deconstruct a text that will be provided to you by the second  collaborator of the event (possibly an academic lecturer about  entrepreneurial/corporate language) in order to unpack its extensive use  of affective language. Upon agreeing to your participation, we can  discuss further and agree together on the exact format of the event and  the level of interactivity with the audience and with the other contributor/lecturer. 
The  event will take place on the 16th of March at WdW, and is now planned  from17:00 to 19:00h (with some flexibility if necessary). Witte de With  will reimburse your travel costs from Brussels to Rotterdam and for your  participation we can offer a moderate fee. 
Because  our "Class of 16" project is somewhat of a pressure-cooker of art  educational collaborative co-programming exploration, we are writing you  on a short notice; for which we hope for your understanding. 
Looking foreword very much to your reply. 
With warm regards, also on behalf of the other students of the Class of '16, 
Lila 


About Class of '16 

This year, the Witte de With Class of ‘16 commenced  as Witte de With Center for Contemporary Art’s brand new initiative for  higher education student co-programming. This educational programme  will form a co-programming team óf and fór students every study year,  aimed at collaboratively exploring art education on the intersection of  Witte de With as art institution and (the curriculums of) partner  institutes from higher education.

Class of '16 is part of Studium Witte de With,  the higher education platform for art and theory by Witte de With  Center for Contemporary Art. Studium Witte de With is intended to serve  as a catalyst bridging various fields of knowledge across higher  education. It presents lectures, debates, workshops, and other  programming in collaboration with partner institutes, rooting the  programme firmly in the curriculums of Academies and Universities. 
??
?

http://lav.io/2014/06/videogrep-automatic-supercuts-with-python/
edit videos based on srt-files
Gijs used videos on youtube-el, when you download the film, comes with srt file
-> take difference between end & start time of film -> video clip
-> is not precise enough, leaves the entrie sentences as subtitles

Proposal 1: construction of database of clips/words based on srt-files, generate new videos
generate new text on selection of videos
https://rg3.github.io/youtube-dl/
sudo pip install --upgrade youtube_dl

Proposal 2: to work with audio recognition using pocketsphinx -> for later!
http://cmusphinx.sourceforge.net/
you can also feed srt-files to Pocketsphinx, find best guest...

Agenda
Focus on 4 words: 'community', 'participation', 'radical future', 'space', 'sharing', 'open', 'storytelling', 'value' , 'solution'
* Selection of videos (Sam, Piero)
http://pad.constantvzw.org/p/startup_vocabulary
* Find srt-files (Manetta) (after updating youtube-dl, it doesn't give an error anymore!)
* Generate srt-files - Use pocketsphinx to generate SRT's (Gijs)
* Text generation : word/sentence tokenizing of text in srt-files (An)
deduct start from end time = length of sentence
string length: put time on letter
calculate at what position it starts + bit time before/after
-> numbers are confusing - rewrite numbers as words
-> weight on punctuation
* Connection/ Interface to database : 
- Corpus of text (collection of words/sentences)
-  Corpus of images
* Generating machine:
    * create collections of videos + srt files - script that scrapes links --> http://pad.constantvzw.org/p/video-sources-links
    * fill db -> central db
    * create script to generate sentences: input sentences, analyze POS, query db based on POS

Examples:
Casetteboy's Cameron rap --> https://www.youtube.com/watch?v=0YBumQHPAeU
example VTT --> http://pad.constantvzw.org/p/writing-with-film_example-vtt

Code:

creating db & parsing srt files in words/sentences:  http://www.algolit.net/scripts/writing-with-film.zip

notes on the writing-with-film.zip installation 


---

NOTE: all these scripts are written in Python2.7 (because of Pattern of Python)

* README file
http://pad.constantvzw.org/p/writing_with_film_README

* sentence-pos-parser.py  
script to translate a sentence into pos-tags .
to use it, run: 
    python sentence-pos-parser.py --text "Write your sentence here."
http://pad.constantvzw.org/p/sentence-pos-tag-parser

* create a database with sentences/words/tags/timecodes from a set of films/videos
    # add film + srt file in input folder 
    # adapt names in script on lines 12 and 13
    # change name of your database if you want on line 73
to use it, run: 
    python fromsrt.py

* generator.py 
script that generates a query to the database based on a set of words (line 44) :
http://pad.constantvzw.org/p/writing_with_film_generator
to use it, run:
    python generator.py
play with framerate line 47:   final.to_videofile(output, fps=29.97) # Export generated clips into video

* tag-generator.py
tagbased generator asking for a sentence as input  ( make sure you first have generated a collection)
to use it, run: 
    python tag-generator.py --text "Write your sentence here."
http://pad.constantvzw.org/p/writing_with_film_tag-generator

* words-generator.py
words-based generator asking for a sentence as input  (make sure you first have generated a collection)
to use it, run: 
    python words-generator.py --text "Write your sentence here."
http://pad.constantvzw.org/p/writing_with_film_words-generator

* software: Youtube dl
Youtube-loader: http://pad.constantvzw.org/p/youtubeloader_writing_with_film
https://github.com/rg3/youtube-dl

* fromvtt.py
a parser for subtitle files in the VTT format:
http://pad.constantvzw.org/p/writing_with_film_subtitle_parser_vtt

--- 

Hyper Audio, http://hyperaud.io/
an open source software package comparable to videogrep
funded by Mozilla a bit, as it alligned qith their ... 

Pocketsphinx:
    script takes video file, analyses audio file, generates words, puts them in db, linked to sample
-> db with start & end time from video you can use to generate new clip

---

Keywords: Community, Participation, Space, Radical Future, Storytelling

https://www.youtube.com/watch?v=wbP6N1P2-Eo
(Superscript 2015 Panel: Connectivity and Community)

https://www.youtube.com/watch?v=6bjBTOt0nJ8
(Become a Stedelijk Member - Tino Sehgal)

https://youtu.be/T8TIdVETbRU?list=PLLrFzV6gBibdNNUHTqj9MNGMjNK6O0tj0
(Julia Peyton-Jones introduces day two of the Extinction Marathon: Visions of the Future )

https://www.youtube.com/watch?v=JHPVPrT4xJk  
(Hito Steyerl on Art as Occupation)

https://www.youtube.com/watch?v=c1FhZqpbgDw  
(The Space)

https://www.youtube.com/watch?v=eYqKSTDpjrY
(Storytelling in the Archives | Performance Forum )


--
Example of SRT
1
00:00:10,243 --> 00:00:11,211
The President: Hi everybody.

2
00:00:11,211 --> 00:00:13,913
It's now been a year since
the tragic death of Michael

3
00:00:13,913 --> 00:00:16,216
Brown in Ferguson, Missouri.

4
00:00:16,216 --> 00:00:19,285
His death-along with the
events in Cleveland, Staten


link to LOTE5 event this week/end: https://edgeryders.eu/en/lote5/welcome-to-lote5
Sam hackaton: https://oscedays.org/