auto-correct a speech-to-text output and relate to of the words based on syllables
On 2018-02-03 09:34:57 +0100, dieter wrote:
> Steven D'Aprano <steve+comp.lang.python at pearwood.info> writes:
> > On Fri, 02 Feb 2018 08:14:03 +0100, dieter wrote:
> >>> The user speaks "Light". The system translates it as "Bright" The user
> >>> speaks "White" The system translates it as "Bright"
> >> As those words are phonetically quite apart (they have very different
> >> first consonants), some step in your processing chain does something
> >> seriously wrong.
> > I disagree: Light, Bright and White sound very similar. They're identical
> > except for the first consonant:
> We have here an example that the first consonant can significantly
> influence the meaning.
> As a consequence, it will in general be spoken and affect the sound.
It will affect the sound, but not the same for every speaker. Since you
have a German mail address, consider how differently "Chemie" is
pronounced in German depending on which part of Germany you are from.
Now consider someone speaking with a French or Indian or Chinese
accent. Different languages have different phonemes, and humans
generally learn to distinguish between them (and to disregard variances)
in the first years.
> And obviously, I should not be ignored when one is interested in
> a narrow match.
The difficulty is to *recognise* it correctly. Was that tangle of sound
waves an "l" or an "r"? This not as unambiguous as you seem to think.
So a speech-to-text program may hear "right" when the speaker was really
saying "light". If you have only the output from that program you must
determine whether "right" is correct or must be corrected to "light".
_ | Peter J. Holzer | we build much bigger, better disasters now
|_|_) | | because we have much more sophisticated
| | | hjp at hjp.at | management tools.
__/ | http://www.hjp.at/ | -- Ross Anderson <https://www.edge.org/>
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 833 bytes
Desc: not available