Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
Tip your server.

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.

user:
pass:
register,


             

Whisper to voice software

For laryngerial cancer
  (+1)
(+1)
  [vote for,
against]

See the link of this medical doctor who had cancer, and gives a speech about it in a whispering voice.

If he and others would have "whisper to voice" software we could listen to their discussions in a much easier way.

It would also be useful for the workspace, and the train ride, where you want to talk privately, and quietly, and to show your consideration for the other people riding with you or working in the nearby cubicle.

pashute, Jul 16 2017

https://www.youtube...watch?v=SegAy6rdBb4 [pashute, Jul 16 2017]

[link]






       No remarks? Will it not work?
pashute, Jul 25 2017
  

       /No remarks?/ The fading of the halfbakery, I am afraid.   

       But this could be done. Voice recognition software now is excellent and some software can learn; for example, a foreign accent. Voice recognition should be able to learn your whisperer and restate his whispers using Microsoft Sam or comparable machine voice.   

       Or he could whisper into a microphone and with added loudening, route that through some stacks of amps borrowed from AC/DC.   

       Side note: I learned that speech recognition could not distinguish 2 different whistled tones.
bungston, Jul 25 2017
  

       Or you could implant sensors to determine the muscle movements of the jaw/tongue/larynx and simulate the vocal tract to determine what was intended to be said, then synthesize it electronically using the same formants and model of the vocal tract. In fact you might be able to cut out the middle man and get those signals directly from the brain.
gtoal, Jul 25 2017
  

       Hi gtoal. Thanks for the remark. I want to make something that just works out of the box with a microphone and no need for any implants.   

       I tried all kinds of existing vocoder affects and none worked. We'll need to have some kind of algorithm, where speech reco would drive the artificial vocoder.
pashute, Jul 27 2017
  

       Maybe look into throat microphones designed to pick up subvocalization. IIRC, the US Army was looking into that a few years ago for silent communications between soldiers.
notexactly, Aug 15 2017
  
      
[annotate]
  


 

back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle