[SLED] Voice and Education in Second Life

Matthew Dowd matthew.dowd at hotmail.co.uk
Sat Aug 4 03:16:52 PDT 2007

I should open this e-mail to say I've no issue with voice per se (since on re-reading this e-mail it sounds pretty negative), just I'm not convinced it is all positive!

The lack of a transcript has been one of the things I've been expressing concern about (mainly on the SL forums), so I'm glad that Terry has empirical evidence to support that (especially as a response on the SL Forums I received was "well you have to take notes in RL so whats the problem taking notes in SL" - which sort of missed the point that we shouldn't be limited in SL by what happens in RL!).

Although as mentioned, there are means of capturing the audio and converting to text, that misses the real time nature of the current text transcript. Let's say I'm listening to a voice lecture in SL and don't understand something said. If I IM my neighbour (even in text), the chances are I'll get a "shush, I'm listening" type response. If the lecture was in text, however, the odds are my neighbour may give a more useful response since they can take their attention off the main lecture briefly knowing they can pick it up again by quickly scanning the history.

This also has relevance as regards the RL environment of the learner - we all know that one of the motivations behind distance learning is to allow the learner to fit learning around all their other concerns, distractions etc. In attending a text based class, I can briefly step outside the class if something crops up in RL, and pick things up quickly from the chat history (this garnered another "but in RL you have to cope with missing things if you step out of the room..." which missed the point).

There are of course other RL environmental issues - not everyone can necessarily closet themselves away where they can speak to their computer without annoying others (or feeling silly). Personally, I've yet to find a headset I can wear for any reasonable length of time.

However, my big bugbear with voice is lack of support in the scripting language. For instance, I've been playing with text of a prim scripts recently, and have created a large noticeboard which displays the last 6 lines spoken within its earshot (including the speakers firstname, and with different colours to represent the speaker). I'm not sure whether it would be of any real use within an educational context, but I'm sure someone might be able to use a variant of it, or something completely different which responds to "speech" around it. However, at present, such ideas and devices are not possible when the medium is voice rather than text. There's been a little discussion on the SL jira on this, but very little interest from Linden Labs themselves, and there is a concern that the fact voice has been outsourced to a third party may make voice related LSL functions difficult to implement architecturally (and of course the voice to text/text to voice issues aren't simple).

Feel like a local wherever you go with BackOfMyHand.com

More information about the Educators mailing list