Call for Participation ...

[percival]

A student of mine has done an M.Sc. thesis on choosing the right voice for smart homes. It was a very nice thesis, and all that prevents us from publishing it is lack of participants, sooo ... cue some more recruiting!

My student has set up a neat, short little experiment here

It will only take you 15 minutes. All you need is a computer with sound and a reasonably up-to-date version of Java (1.5 or better), and off you go. I'll post results to once we've analysed the additional data.

If you enjoyed this little experiment, feel free to link to this post from your own journals. I'll keep this post up until the end of November, then it will go private.

If you have any problems or if you would like to know more about the experiment, just comment.

In case you are wondering about the gift certificates - we're going to chip in with some more £15/$25 amazon certificates as an added incentive ;)

Comments

[lizr0221.livejournal.com]

I will do this tonight :)

[perceval.livejournal.com]

Thanks :)

[sidheag.livejournal.com]

Here's what I just mailed to the email address on the web page :-(

I tried to do your experiment, and everything appeared to work, until I
clicked the Play button when nothing happened. Sound was turned right up,
and the sound on my computer works in other contexts. I was running Mozilla
Firefox version 2.0.0.9 under Windows Vista, with JRE 1.6.0. Here's the
content of my Java console.

Java Plug-in 1.6.0_02
Using JRE version 1.6.0_02 Java HotSpot(TM) Client VM
User home directory = C:\Users\perdita


----------------------------------------------------
c: clear console window
f: finalize objects on finalization queue
g: garbage collect
h: display this help message
l: dump classloader list
m: print memory usage
o: trigger logging
p: reload proxy configuration
q: hide console
r: reload policy configuration
s: dump system and deployment properties
t: dump thread list
v: dump thread stack
x: clear classloader cache
0-5: set trace level to
----------------------------------------------------

INFO Trying server: fordyce.inf.ed.ac.uk 9876
INFO SUBJECT sent
INFO Finished reading Timeline with 5 stages and
22 slides.
WARNING Could not get html parameter feedback
INFO Full screen not supported; using browser window.
INFO Starting play ...
INFO Reading audio clip from
http://fordyce.inf.ed.ac.uk/webexp2/resources/smart_audio/TEST_garagedoor_neg_happy.wav
UTF8--**-L6 C0 Top mail to greta.boye.experiment@googlemail.com 12:50pm (Mail Fill Abbrev)------------

[perceval.livejournal.com]

Thanks! I've forwarded this to the programmer who maintains the experiment environment as a potential bug/problem.

[if0x.livejournal.com]

OK, I can't do this from work because the work PC has no sound (woes), but I'll try to run it from home tonight :D

[perceval.livejournal.com]

Thanks!

[mincot.livejournal.com]

Will do this evening! Sounds like a very good idea.

[perceval.livejournal.com]

Thanks!

Don't read this comment if you haven't done the experiment yet!

[dianora.livejournal.com]

Cool! I just did it.

Do you know why she chose male voices only? I think I would have rated a female voice consistently higher -- but I can't be sure. I thought the deeper male voice was often downright creepy. When he said things like "The security company has not been called" and "The house alarm has not been turned on," I was afraid he was going to kill me in my own home!

Re: Don't read this comment if you haven't done the experiment yet!

[perceval.livejournal.com]

Heee. The manipulation she used had been done for this particular voice only; creating a female voice would have meant too much hassle for the person making the voices.

Thanks for your comment!

[awelkin.livejournal.com]

done.

Catherine

[perceval.livejournal.com]

Thank you! :)

[queenriley.livejournal.com]

Done!

[perceval.livejournal.com]

Thanks :)