Speech recognition today doesn't really work in noisy environments.
Humans are still much better than computers at recognizing speech.
In the speech sound wave, one word runs into the next seamlessly; there are no little silences between spoken words the way there are white spaces between written words. We simply hallucinate word boundaries when we reach the end of a stretch of sound that matches some entry in our mental dictionary.
I can no longer type, so I use TalkingPoint and Dragon Dictate. It's a speech-to-text program, and there's an add-on for talking which some guys came up with.
Speech is not a means in the service of an external end. It contains its own rule of usage, ethics, and view of the world, as a gesture sometimes bears the whole truth about a man.
In order to understand one person speaking to us, we need to process 60 bits of information per second.
I'm very interested in voice synthesis and vocoders in general.
The areas of the brain that have to do with speech are very connected with the same parallel processors that have to do with the kind of ballistic calculations you need to hit small game with a rock.
That's called a microphone. It's a big sausage that picks up everything you say - and you're starting early.
Now the identification of individual sounds by phonetic observation is an artificial way of proceeding.
