Terry Pratchett Quotes

I'm very interested in voice synthesis and vocoders in general.

Humans are still much better than computers at recognizing speech.

Why on earth should I care whether people read me with their eyes or their ears?

The voice I use is a very old hardware speech synthesizer made in 1986. I keep it because I have not heard a voice I like better and because I have identified with it.

I have this peculiar ability to be able to anticipate mouth movements on screen and fill them with words or sound.

I want to be able to speak with errors in my wording, errors in my grammar. When you type things into Google search, it corrects your words. With speech, I want it to be general enough, smart enough, to know 'No, he couldn't have meant these words that I think he said. He must have really meant something similar.'

I remember the beginnings of the Kurzweil reading machine. I was one of the first to meet Ray Kurzweil and purchase the reading machine in Boston. To think that the machine was at least two and a half large suitcases at the time, and now you have a camera and it takes a picture and you have sound.

Speech recognition today doesn't really work in noisy environments.

Around my neighborhood, I'm known as the American who talks to her computer while she types.

It's quite fun to mess with the human voice. It's quite special in the sense that the voice is the #1 instrument that we can connect with; it doesn't sound too alien. I think that's the key is to find the line between sounding human and sounding robotic. That's an area that I like to explore a lot.

There are endless consumer applications, but what excites me is how this can help people. A man who cannot speak communicates with sign language, but the average person doesn't know that language. SixthSense, if equipped with speakers, can recognize the gestures and form the words - it will speak for him.

So I'm reading a book on my new iPad, but can't the iPad read it for me? Do I have to do everything?

That's called a microphone. It's a big sausage that picks up everything you say - and you're starting early.

But I'm so slow on it because I find it terribly hard writing blind on computers. The computer speaks to me, but it's just so slow, I'm so terribly slow using it.

Continuing on my theme of backing inclusive innovations, I am optimistic of the success of Uniphore. Man-machine communication is one of the more complex problems to solve. Uniphore's vision lends possibility of finding a solution to this very difficult problem, and the company has already made substantial progress.

When I was building robots in the early 1990s, the problems of voice recognition, image understanding, VOIP, even touchscreen technologies - these were robotics problems.

Acoustic phonetics, which is developing and increasing in richness very rapidly, already enables us to solve many of the mysteries of sound, mysteries which motor phonetics could not even begin to solve.

Speech sounds cannot be understood, delimited, classified and explained except in the light of the tasks which they perform in language.

There needs to be an app that edits what I say versus what I want to say.

Why did Google, for example, recently decide to offer free 411 service? I haven't talked to people at Google, but it's pretty clear to me why. It's because of speech recognition. It has nothing to do with 411 service: it has to do with getting a database of voices, so they don't have to license speech technology from Nuance or someone else.

Google, as usual, is one step ahead of everyone and provided the means where all videos on YouTube can be automatically captioned through voice-recognition technology without having to be told that it's the responsible thing to do.

I've discovered this new electronic technique that creates new speech out of stuff that's already there.

The issue in Web accessibility is the fact that blind and visually-impaired people need the single biggest boost to achieve equivalence, since the real-world Web is a visual medium.

Big companies, which spend tens of billions of dollars annually on 'call centers' to take orders and provide customer support, increasingly rely on speech recognition not just to handle requests for information but to process customer orders.

You speak into it and everything is recorded, voice, tone, intonation, everything. You turn a little wheel, and forth it comes, and can be repeated ten thousand times. Only fancy what this suggests.