If anyone was still speculating about what my geeky excitement was about last week…
Last week I had a stray sighting of an advert for the Euan Macdonald Centre for Motor Neurone Disease Research. I think it may have been linked on twitter or a blog somewhere: I can’t figure out exactly where I found it. Anyway, it really caught my attention.
What it was asking for was for volunteers to come along and donate their voices for Motor Neurone Disease and similar conditions, which cause people to lose their own voices. Artificially generated voices can be rather robotic. I guess the obvious one is Stephen Hawking, although I understand he has turned down a more natural one, so strongly has his voice been identified with him. The idea of the Voice Bank project is that a bunch of volunteers come and record their voices and those voices can be chopped up and rearranged into something as similar as possible to the recipient’s original voice.
Now, I have been known to rabbit on in my time, so I reckoned I could go and sit in a room and talk nonsense for an hour (a paraphrase of their requirement).
I filled in the forms online with a little difficulty. I’m always a bit challenged by the question “what accent do you have?” Hmmmm. You tell me. I typically surprise people a lot when I tell them I come from (nearly) Paisley and that I lived there for all the first 18 years of my life. I do not have a typical Paisley accent. However, in the absence of a box for “grew up near Paisley, honestly, my mum’s accent was Glasgow with a wee bit of Derbyshire and she didn’t like me speaking too broadly and yes I know you think you can hear something non-Scottish in there but I don’t know what it is either”, I ticked the Renfrewshire box. And signed up for my one-hour appointment.
It was all rather straightforward. I was ushered into a basement office where the basic stuff was explained again and I signed a waiver form to say I wouldn’t claim any performing rights money!
Then I went into an anechoic chamber where I sat on a rather snazzy Panton chair and put on a big set of headphones with a microphone. The researcher who was running the recording could talk to me through the headphones and also watch me on her computer screen (something I only remembered halfway through. I hope I wasn’t scratching my arse.)
I had my own big screen in front of me where the words came up for me to read. The sentences started off by telling coherent stories – the first sentences were something American-sounding about going to the store, then there was a little science story – but got more and more unrelated.
I did wonder whether I’d feel claustrophobic in the anechoic chamber (it was quite small, and closed) but it was nice and bright and no problem at all.
There was a break after every 50 sentences for a drink and a stretch. It’s a long time since I talked non-stop for nearly an hour (some people may beg to disagree on that) and you do get quite dry!
I was asked to talk in as normal a voice as possible and not be too stagey. It was quite weird at the start. Normally even if I was speaking in public, I wouldn’t be using words that are all written down, just prompts. Here, if I mistook a word, we had to go back to the start of the sentence. I found it particularly challenging to say “It is” every time instead of “It’s” for example. It was hard to make that sound natural.
It was quite interesting being in this little room with only my voice for company. I’m aware of how much I change my accent depending on who I’m speaking to (not usually consciously) so it was interesting to hear myself settle down into something close to my real voice. The one that I use when I’m alone. I started off in a bit of a public/reading out loud voice, but relaxed into something a little more Scottish.
The constant flow of words in front of me was weirdly meditative. Each sentence came out of the blue. Odd little images kept popping into my mind. Occasionally I thought they were getting at me (“I have no maternal instincts.” “I am now unemployed.”). All the while noticing: hmm, I almost drop the t at the end of at and it, but I don’t really have a glottal stop in the middle of words. I do say ‘I’ a bit like ‘Ah’ – definitely the West of Scotland coming out – but I don’t say ‘was’ like ‘wiz’. Some of my vowels are a lot more Scottish than others.
I did get a bit uncomfortable about 2/3 of the way through, but that’s just me and chairs (I’m sure if they’d let me sit on the floor I would have been fine). I could stop whenever I wanted and only got told off once for shuffling while reading!
Apparently I’m a “good reader” and stumbled less than most people, although I did have the odd tongue twister! I’m always tempted to read how I think the sentence should go rather than how it actually goes.
One assurance that I had in the forms I filled in was that nobody would be given my voice as a whole. In fact, I was told that even the individual sounds are averaged over many speakers. I really wouldn’t mind if anyone had my voice, but I can’t imagine that they’d want it. I grew up sounding so similar to my sisters and mother that no-one could tell us apart on the phone. I was thinking, even if I lost my voice, I could just get them to record a sister’s…
I do find my voice frustrating sometimes: when I hear it recorded it’s too high-pitched and young-sounding. When I get a taxi in Renfrewshire they never believe I’m local. I have to go through the accent guessing-game with Scots and then have Americans be surprised that they can actually understand me.
But it sounded nicer in that deadened room. And I just geekily love the idea, that someone somewhere could be going around with my Ks or As or Ts and never know it. They’d just know that they’d got something approaching their own voice back.
If anyone else is interested, the MND Voice Bank has had a good response but is still taking volunteers. Men would be particularly welcome. There are some evening appointments available as well as during the day.