I became interested in text-to-speech and voice recognition recently.  Partly because I started recording voice notes on my BlackBerry Bold recently, instead of adding new todo items and typing out text memos of new ideas.

Back in 2003, I pondered using text-to-speech to perform intelligent “morning buddy” tasks.  Originally it was motivated by a post on Seth Godin’s blog,  In other words – how would you like somebody to wake you up (and I mean persistently bother you until you actually get out of bed); and somebody to read you the morning news, weather and todo list; and what if that thing was your computer – or even a little bed-side appliance?

Over the last couple of days I started thinking about this idea again.  Originally I thought it would be very cool to control your instant messenger application with voice.  Think about driving and getting an IM alert from your phone.  Unlocking your phone (as I use the BlackBerry password lock feature religiously) with voice would be cool.  Then you could have the phone read the IM message to you.  Prompt it to read you the previous message to give you context of the conversation – and it does that.  Prompt it to reply and it uses voice recognition to type a response and then read it back to you.  Tell it to send the message when you’re ready, and presto! – voice-based instant mobile messaging.  I’ve used the Voice Command feature on Windows-based mobile phones like the Treo.  I know this kind of thing can be done.  But if somebody created something like this already, why don’t I know about it? Tell me if I’m wrong, I’d love to just use something like this that’s already built.  But somehow I think its either not a reality yet or perhaps it hasn’t been properly marketed.

Either way, I’m going to make this a little pet project, or something like this.  I have no doubt that there’s a demand for something like this.  Now I’m gonna have to get my google search on, and then get my hands a little dirty with code to make it happen.