Wednesday, February 22, 2006

Wired for Speech

I'd just gotten out of a lecture that was partly about how fine tuned our minds are to detecting human speech, picking it out from ambient sounds, etc. I sat down about 20m from a pair of guys, just close enough to hear that they were speaking. I couldn't actually make out any of their words, but it sounded like Swedish. I don't know how to describe that, whether it was which syllables were emphasized or the pace of the conversation or what, but I listened closer and finally heard a word that possibly could've been Swedish. So I walked up to them after finishing my lunch, hoping I was correct or I'd look like a fool, and sure enough, they were from Sweden. This wasn't the first time something like this had happened. Just as our minds are easily able to pick out human speech, is the same true for distinguishing languages of human speech?

I saw the owner of this bike on my way to campus this morning. The iPod was blaring in full effect, coasting down the bike lane.

Update: The professor of the above lecture said today that we are able to distinguish our native language from other languages when we turn 4 days old. Remarkable.


Alex said...

I think you're right -- I think we are somehow able to pick out languages. When I was in France, I got pretty good at picking out languages around me when we were travelling. I didn't even have to know the language to be able to pick it out; I just had to have heard it before.

Also, it seems like once we learn a language, our brain stops feeding us information about the sounds and starts feeding us information about the words -- it acts as a tokenizer. This makes more sense when you think about how a word starts to sound strange when you say it too many times in a row...

Just some random thoughts.

didier said...

For me, identifying languages involves 2 things: the rythm and the tone (like you mentioned). It seems that every language has its own rythm that sets them apart.