User:Max Dovey/Tuesday18thfebuary: Difference between revisions

From XPUB & Lens-Based wiki
(Created page with "http://headroom.pzwart.wdka.hro.nl/~max/fulloutput.mp4")
 
No edit summary
 
(11 intermediate revisions by the same user not shown)
Line 1: Line 1:
===Tuesday 18th February 2014===
I recorded myself using this microphone badge i put together. The recording (10 hours) was then put through speech to text and the whole day is transcribed. The Result is a 10 hour video (just sound) with the subtitles interpreting the sounds as you hear them.
In this experiment the range & capabilities of the speech translation program (BMP adobe premier pro) is tested and its a response to phonology and acoustics is understood.  <br>
[[File:New2.jpg|left]]
[[File:Badgemic.jpg|left]]
<br>
{{youtube |SLP21afOI6E }}
Adobe After effects Subtitles Script.
For use with adobe premiere's transcription
<source lang = "javascript"
L = thisComp.layer("name of your video. extension");
max = 5; // number of words to display
n = 0
  if (L.marker.numKeys > 0){
    n = L.marker.nearestKey(time).index;
    if (L.marker.key(n).time > time){
      n--;
    }
  }
s = " ";
if (n > 0){
  base = Math.floor((n-1)/max)*max + 1;
  for (i = base; i < base + max; i++){
    if (i <= L.marker.numKeys){
      s += L.marker.key(i).comment + " ";
    }
  }
}
s
</source>
link below
http://headroom.pzwart.wdka.hro.nl/~max/fulloutput.mp4
http://headroom.pzwart.wdka.hro.nl/~max/fulloutput.mp4

Latest revision as of 20:52, 26 March 2014

Tuesday 18th February 2014

I recorded myself using this microphone badge i put together. The recording (10 hours) was then put through speech to text and the whole day is transcribed. The Result is a 10 hour video (just sound) with the subtitles interpreting the sounds as you hear them. In this experiment the range & capabilities of the speech translation program (BMP adobe premier pro) is tested and its a response to phonology and acoustics is understood.

New2.jpg
Badgemic.jpg




Adobe After effects Subtitles Script.

For use with adobe premiere's transcription
 0){
    n = L.marker.nearestKey(time).index;
    if (L.marker.key(n).time > time){
      n--;
    }
  }
s = " ";
if (n > 0){
  base = Math.floor((n-1)/max)*max + 1;
  for (i = base; i < base + max; i++){
    if (i <= L.marker.numKeys){
      s += L.marker.key(i).comment + " ";
    }
  }
}
s

link below http://headroom.pzwart.wdka.hro.nl/~max/fulloutput.mp4