Searle's Room

From Robert-Depot
Revision as of 14:06, 28 May 2013 by Rtwomey (talk | contribs) (GRBL settings)

Jump to: navigation, search

Searle's Paper

Searle, J.R., "Minds, brains, and programs," The Behavioral and Brain Sciences, 3, 1980.

Rooms

  • Ernest Hemingway. A Clean Well-Lighted Place. 1933.
  • William Pope.L. White Room #4 / Wittgenstein & my Brother Frank. 2005. [1]
  • Jorgen Leth. The Perfect Human. [2]

Context

Dear Gallery Goer,
This is a press release for the first exhibition of art work by my son, William Pope.L,
at Catherine Bastide Gallery in Brussels, Belgium.

The works in this exhibition borrow from many media: pink liquid antacid, changing
stations, miniatures of monuments to the future, mattresses, children's clothing,
fish tanks, drawings and video. Most of the work was made in 2008 but a few special
works date from 1998 and 2002.

For example, my son believes that ideas such as 'childhood' and 'future' are 
sentimentalized. He believes that we valorize these concepts by emptying them of
uncertainty. He believes the real power of these ideas are their inherent emptiness,
not the emptiness we impose. The emptiness we impose is cheap, polite, and 
apolitical.

I do not agree with my son but he is my son and so I recommend him without reservation.

William Pope, father of the artist
Chicago, Illinois, USA
January 21, 2008

Translation

  • The New Translation Manifesto. [19]
    • an ever-expanding browser which seems to depict itself...is more beautiful than Parmigianino’s Self-Portrait in a Convex Mirror.
  • The Futurist Manifesto [20]
  • Walter Benjamin. Task of the Translator.

Technical

Tesseract

Recognition

configuration

Train A Language Model

http://www.speech.cs.cmu.edu/tools/lmtool-new.html

interpreting baby speech

  • resample audio
    • sndfile-resample -to 16000 trimmed3.wav resampled3.wav
  • default model
    • ~/supercollider/solipsist/binaries/pocketsphinx-osx-continuous -infile resampled3.wav
  • new 70K model:
    • ~/supercollider/solipsist/binaries/pocketsphinx-osx-continuous -infile resampled3.wav -lm /usr/local/share/pocketsphinx/model/lm/en_US/en-70k-0.1.lm
  • with a grammar of 15000kY/old ultraconserved words (very slow):
    • ~/supercollider/solipsist/binaries/pocketsphinx-osx-continuous -infile resampled3.wav -jsgf /Users/rtwomey/Documents/Classes/dxarts_general_exam_2013/ultra.gram
  • with a trained small lm of 15000K old ultraconserved words:
    • ~/supercollider/solipsist/binaries/pocketsphinx-osx-continuous -infile resampled3.wav -lm /Users/rtwomey/Documents/Classes/dxarts_general_exam_2013/ultra.lm -dict /Users/rtwomey/Documents/Classes/dxarts_general_exam_2013/ultra.dic

Shot Detection

  • run a shot detection on a movie
    • ./shotdetect-cmd -w -v -f -l -m -r -s 60 -i /Volumes/Cistern/Memory/home\ movies/Old\ Family\ Home\ Movies\ 2\ of\ 21-HrcOEqnOAQ0.flv -o ./ofhm2/

CHILDES speech database

CHILDES is the child language component of the TalkBank system. TalkBank is a system for sharing and studying conversational interactions. http://childes.talkbank.org

Processing

  • download mp3 audio from database
  • trim a child speech utterance out of a wav file
    • sox wav/1-05-09.wav speech1.wav trim 3.901 4.528
  • extract and concatenate all child speech from a transcript and wav file:
    • python ~/code/pytwomey/child_speech/extract_child_speech.py ./data/1-05-10.cha ./wav/1-05-10.wav ./processed/1-05-10/
  • extract audio from video documented file:
    • ffmpeg -i af06.mov -vn -ac 2 -ar 48000 af06.wav
  • extract audio of child directed speech from file:
    • python ~/code/pytwomey/child_speech/child_directed_speech.py ./data/c1/c1-0917.cha ./wav/c1-t14jan97.wav *MOT

Subsets

  • Brent.This corpus includes about 100 hours of recordings from English-speaking mothers of preverbal children. All of the transcripts are completely linked to the audio. [21]

FFPLAY

Children's Drawings

Children Writing

process

  • upsample image
    • convert /Volumes/Cistern/Memory/childrens_drawings/spellingphase11.bmp -interpolate bicubic -resize 4000x3000 +repage out.png
  • potrace, output as geojson
  • import with geojson 1.0 python

skeletonization

GRBL settings

Program all at once

  • open screen:
    • screen /dev/tty.usbmodem1421
    • copy and past the following (repeat any lines that fail with "unsupported statement"):
$N0=G20
$N1=G91M3
$0=6.964 (x, step/mm)
$1=6.964 (y, step/mm)
$2=6.964 (z, step/mm)
$3=10 (step pulse, usec)
$4=20320.000 (default feed, mm/min)
$5=65000.000 (default seek, mm/min)
$6=56 (step port invert mask, int:00111000)
$7=25 (step idle delay, msec)
$8=900.000 (acceleration, mm/sec2)
$9=0.050 (junction deviation, mm)
$10=0.100 (arc, mm/segment)
$11=25 (n-arc correction, int)
$12=3 (n-decimals, int)
$13=1 (report inches, bool)
$14=1 (auto start, bool)
$15=0 (invert step enable, bool)
$16=0 (hard limits, bool)
$17=0 (homing cycle, bool)
$18=0 (homing dir invert mask, int:00000000)
$19=25.000 (homing feed, mm/min)
$20=250.000 (homing seek, mm/min)
$21=100 (homing debounce, msec)
$22=1.000 (homing pull-off, mm)
ok
  • Check GRBL settings
  • You should see:
Grbl 0.8c ['$' for help]
G20ok
G91M3ok
$0=6.964 (x, step/mm)
$1=6.964 (y, step/mm)
$2=6.964 (z, step/mm)
$3=10 (step pulse, usec)
$4=20320.000 (default feed, mm/min)
$5=65000.000 (default seek, mm/min)
$6=56 (step port invert mask, int:00111000)
$7=25 (step idle delay, msec)
$8=900.000 (acceleration, mm/sec^2)
$9=0.050 (junction deviation, mm)
$10=0.100 (arc, mm/segment)
$11=25 (n-arc correction, int)
$12=3 (n-decimals, int)
$13=1 (report inches, bool)
$14=1 (auto start, bool)
$15=0 (invert step enable, bool)
$16=0 (hard limits, bool)
$17=0 (homing cycle, bool)
$18=0 (homing dir invert mask, int:00000000)
$19=25.000 (homing feed, mm/min)
$20=250.000 (homing seek, mm/min)
$21=100 (homing debounce, msec)
$22=1.000 (homing pull-off, mm)
ok

Moving Image/Video Archives