h a l f b a k e r yTip your server.
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,
|
|
|
ShazamTalk
Music Recognition Aides Voice Recognition | |
Let's face it -- even with the latest advances in voice
recognition, dictating actual messages can be a lengthy,
error prone process, especially when you're driving.
Yet music recognition works surprisingly well, often
needing just a second or so of audio to recognize a tune.
Said tune,
of course, has lyrics.
The idea then is to create use music recognition in
aiding
voice recognition. To create a message, one would sing
a
verse or even a few words from a song on an appropriate
subject with a suitable subset of lyrics.
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Annotation:
|
| |
I experimented with a voice recognition program to
see if it could distinguish two differently toned
whistles. It could not. I conclude that voice
recognition programs are all about rhythm. |
|
| |
[Bung], the voice recognition program didn't recognize the
difference because you were whistling, not vocalizing. It
simply disregarded the sounds. |
|
| |
This would work well towards morphing English into a tonal language. |
|
| |
If this takes off in a big way, there would presumably
be a drive to make popular songs more relevant to
the dictation needs of the busy executive. |
|
| |
Not, of course, that I would condone Halfbakers
listing song titles adapted to such needs. |
|
| |
That's three more points on your poetic license,
[MB].
Any more of this and you'll have to take a remedial
course in paralepsis parking. |
|
| |