h a l f b a k e r yThere's no money in it.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
I was using my cell phone which is capable of
transferring speech into text. It does it so fast, so I
had a thought why not make electronic ear plugs
that filter out messages of a particular type pre-
specified by the user.
The user would select what they did not want to
hear.
For example
a Mommy might set the system to not let her hear
a repetitious "Mommy"
A politician might set the system to filter out any
statement that can be defined as a question.
An overly sensitive individual may not want to hear
talk about snakes or maggots eating peoples eyes.
[link]
|
| |
The trouble is, that no matter how quick the system is, it has to hear the offending word before it can recognise it and block it. And if the user is listening with their ears, they will also hear the offending word as it is spoken but before it can be blocked. |
|
| |
One way that you could implement this, however, is if the user wears noise-blocking headphones which cut out all external noise. Then you can relay the sounds from around the user with a delay of, say, 1 second. (the delay should be set to the length of the longest string to be blocked). Then, the system can hear the offending word, and switch to silence, or “beeeeep” or “mmm, delicious” or “Sponsored by Campbelbury's Condiments for your Seasoning Satisfaction” or whatever the user chooses, before relaying the edited audio stream to the earphones. |
|
| |
OK, the 1 or 2 second delay might make your social interactions a little stilted and awkward, but I'm sure you would acclimatise to it pretty quickly - people manage transatlantic Skype conversations no problem. |
|
| |
what if it had a future predicting algorithm that magically knows and preemptively blocks the words, with no lag? |
|
| |
Yes, that would be much better. |
|
| |