User:Themsen/STT/2

From XPUB & Lens-Based wiki
< User:Themsen
Revision as of 21:54, 28 April 2015 by Themsen (talk | contribs) (→‎Barendt & Michael)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Barendt & Michael

  • Dictation software
  • Not really experienced in coding
  • Focus on one aspect, don't string together different things
  • Try different engines, (speech-to-text)
  • Different software works differently
  • Exploit why it fails
  • Girls playing with speech to text: ended up with 'New York Pizza'
  • Make a selection with what you want to do with this
  • Topic: Miscommunication of the digital
  • What is inherent in digital & miscommunication?
  • Focus:Speech to Text
  • Barendt: how are you going to get me excited?
  • collect funny mistranslations
  • What is on the backend
  • Good forms, try different speech-to-text
  • Automatic Youtube comment-section
  • Different interpretations from speech-to-text
  • Talk to max (he's been into speech-to-text)(Uses Dragon Naturallyspeaking)

Max

  • Google Webspeech unstable because it uses Google API, can be removed at any time
  • Google might take it away as it owns it
  • (Dragon) Record your audio into .flac file
  • Google Webspeech API, people who have hacked it
  • Download Dragon NaturallySpeaking
  • Adobe Premier comes with speech analysis engine
  • Don't go into speech-to-text code, focus on interaction