Sep 21, 2016

Ok Google and Siri

Google speech-to-text logoI like talking to my phone, specifically using voice to text software.

While driving, I can speak a text to my sister and Google will have it (most times) word for word correct - punctuation aside. When I just finished reading a story on my phone, and don't want to review it on my dinky keyboard, I can say it out loud and then add the periods. Picking out my clothes in the morning, I ask my phone about the weather and it tells me it while I'm moving hangers.

Recently, I started getting back into using DuoLingo - a foreign language teaching app. Some of the exercises include saying phrases out loud to your microphone and the app checks your pronunciation. I'm not learning Spanish from scratch - I studied it in high school and uni - and I've had teachers from different areas. My accent is a bit mixed.

So sometimes it's frustrating when the app thinks I said something wrong when I know I didn't. The fix is to usually slow down my speech and really enunciate, speaking clearly and struggling to really pronounce my 't's.

(Here in the Midwest, our "ts" often sound like "ds". I remember talking to a foreigner about a waterfall, and he had to stop me to ask what 'wader' was.)

tap to speak on Samsung Galaxy phone
Yes, Google capitalizes first letters.
Once I noticed it with the DuoLingo, I became aware of all my speech interactions with JARVIS. I use the same adjustments to my speech, though less extreme, when talking out texts.

Personally, I think this is fascinating. I'm changing my speech pattern for my phone. I speak slowly. I make sure to distinguish between sounds. I pause longer between sentences, to break up 'phrases' as google understands them. Sometimes, I even change my vocabulary to avoid longer words I think my phone will have problems with.

Deep into my edits of Stars, and the prevalence of computers in this story, I can't help but wonder if, as we talk to computers more and more, human speech patterns evolve to make those conversations easier to comprehend. And what those changes are.

A reduction in vocabulary? More small word phrases? A loss of rambling? Slower conversations? Will we speak punctuation marks? "Are we still on today - question mark - tell me by noon - full stop"

I'm curious - do you guys do this too? Notice and change your language as you interact with technology? If so, what? 


  1. I have Cortana on my phone, but I haven't been brave enough to try it. I've never tried dictating prose either, because something really bizarre happens between my head and my mouth - those magical words in my brain sound so mundane and pointless when I speak out loud. I remember having a dictafone a long time ago, but hating the sound of my voice too much to use it very often!

  2. Haha! I've done recordings of stories before and I'm not super fond of my voice either. It never sounds like me. But yeah, I've never dictated a story and I don't think I could. Something about the physical process of writing/typing is great for idea generation. That, and quick refreshers of what happened a page ago. XD