So I’ve had a thought about the direction of language learning.  I’ve been experimenting with 3D interactive worlds (specifically OpenSim), which are programs which let people interact with each other real-time in 3D.  Pretty cool stuff.  This is already being used to help people learn languages as many of the 3D servers offer the ability to communicate with each other via voice and text.

There are a couple of problems I can see with doing this activity with students in a class.  The most important problem is that it is extremely difficult (or expensive) to find a real human being that speaks the language you want to know and who has the time to interact with your students.  It can be incredibly difficult to find an entire classroom’s worth of people willing to interact with your students one on one.  Certainly the online nature of the 3d world makes this easier to manage, but still it almost certainly does not happen in most classrooms (although some languages teachers are adopting Skype successfully).

The second problem is the lack of control you have over what happens during the conversation.  Unless your language learners are somewhat advanced, they will probably struggle to communicate effectively with a native speaker, especially early on in their learning.  You also don’t know if they will cover the content you want to cover, or if their conversation even becomes completely off topic or even inappropriate!

Technology has come a long way recently. There are already chat programs which do reasonably well in conversation, especially if they are limited to a specific known topic area.  3d animation is amazing with highly realistic facial animation and human-like gestures, just check out the movie Avatar.  Voice recognition is improving in leaps and bounds every month with some big players (like Google and Microsoft) putting a lot of money into development and again this technology works better when the bounds of the conversation are known.

Imagine we combined these three technologies together.  Students could then be lead through a carefully arranged conversation including, most importantly, the context of the conversation.  All of the subtle cues and body language we use to learn languages can be programmed into the simulations so that students get as close to a real life experience as possible.  Programs of study could be designed for all levels of language learners, allowing for extremely differentiated and customizable instruction for every student in your class.  Instead of having to carefully plan an online session, your students could interact any time from any computer with sufficient power to run the program.  It would also be a fair bit of a fun for the students and hopefully end up engaging them at a deeper level than revision exercises from a textbook.

What’s amazing is that the technologies to implement this are very close to becoming a reality.  Within 2 or 3 years, all of the component technologies to make this work will be mature enough to produce a software package which is stable enough to release into the classroom environment.  Hopefully with a little bit of work on the administrator’s interface, a typical non-techy language teacher could set up their own simulations for their students.

The future language learning looks very bright, although maybe in 10 years we won’t need to learn other languages because the technology will be so advanced that all of our phones will include universal translators.

 

Facebooktwitterredditpinterestlinkedinmail