Mailinglist Archive: opensuse (2532 mails)

< Previous Next >
Re: [opensuse] Speech-to-Text
  • From: "John Andersen" <jsamyth@xxxxxxxxx>
  • Date: Sat, 3 May 2008 19:21:32 -0700
  • Message-id: <60fb01490805031921l36fff6deqafa7942d800c1389@xxxxxxxxxxxxxx>
On Sat, May 3, 2008 at 6:37 PM, David C. Rankin
<drankinatty@xxxxxxxxxxxxxxxxxx> wrote:
John Andersen wrote:

On Sat, May 3, 2008 at 1:23 PM, David C. Rankin
<drankinatty@xxxxxxxxxxxxxxxxxx> wrote:

With all speech recognition software you have to "train" it
which is a fairly time consuming process.

I can't speak to Dragon products, but the need to train is the key
thing that newer
technology is trying to eliminate in speech recognition. My phone
requires no
training, I can just say "call David Rankin" and it would find you in the
address book. Same with Ford Active Sync. Same with some modern
telephone answering systems. (Call HP tech support to hear a demo.)

Admittedly these things are using a vastly reduced vocabulary. But never
less, training is on the way out.

I definitely need a new phone. The Motorola v3xx is pretty good, but
I end up in shouting matches with it just trying to get it to dial "Don".
You know the scenario.

me: "name dial" "Don"
phone: "did you say home?"
me: "no"
phone: "say a command"
me: "name dial"
phone: "say the name"
me: "Don"
phone: "did you say home?"
me: "NO you %$#@#^%#$ phone, I said DON"
phone: "say a command"

Exactly the same problem here with a friend named Dan.
I had to put him in the phonebook as Daniel.
(Moto Krazr k1). This will probably be my last Motorola phone
unless stunning new designs are released, because its getting
to be a fairly old design.

To unsubscribe, e-mail: opensuse+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: opensuse+help@xxxxxxxxxxxx

< Previous Next >