October 12, 2014

His Master’s Voice


Time was, each of my books had a very specific, very carefully chosen soundtrack, a single piece of music I listened to endlessly while working on a book in order to establish in my head a certain tone and rhythm. Hope was that tone and rhythm in my head would somehow bleed out onto the page. Some of the choices were obvious from the start, and sometimes it took longer to find that perfect bit of music than it did to actually write the damn book. So for The Buzzing it was Akira Ifukube, who’d composed the scores for all those classic Godzilla films. For the crime novel Noogie’s Time to Shine it was David Shire’s early seventies brass-and-percussion soundtrack for The Taking of Pelham One Two Three. I chose some early Philip Glass serial music for Unplugging Philco thinking the cold, repetitive music would nicely reflect a cold and repetitive world. And for a variety of reasons, These Children Who Come at You With Knives was written to Bobby Beausoleil’s experimental electronic accompaniment to Kenneth Anger’s Lucifer Rising.

            I don’t know whether or not the technique had any discernible effect, if it actually did make it onto the page, or if it helped me write any better, but it kept me focused and made me happy.

            Then things changed.

            Back then as my eyesight dwindled away toward the inevitable poof, I used a variety of tricks to keep track of what was popping up on the screen. I blew up the fonts to elephantine proportions, switched to a cursor the size of my thumb, increased the contrast, leaned in real close and eventually blew my back as a result. The important thing was my typing skills were solid enough and my guesses good enough that I could get by in a reasonably normal fashion. But then the screen faded to a vague white blur of milk no matter how big the fonts or my cursor, and my typing skills, maybe as a result, took a precipitous nosedive. My once-confident guesswork began failing too, and more and more often I’d find myself ensnared by some computer hijinks I could neither comprehend nor escape after clicking on the wrong button. It was clear I had to make the full-time plunge into the world of electronic screen reading software. So as usual and once again I called Morgan and asked her to help with this, as I (obviously) couldn’t see what the fuck I was doing.

            If I had been unfortunate enough to own a PC, I would’ve had to shell out close to three hundred smackeroos for Jaws, a screen-reading program that clearly and loudly announces every word, every keystroke, every button and whosits and link as I tap the keyboard or swoop the cursor around the screen. I’m told it works quite well, but I didn’t have a PC, and didn’t have the moolah to go dropping it on necessities like that. No, lucky for me I had an Alzheimer’s case of a Mac, and VoiceOver, a program that does exactly the same thing as Jaws, comes built in. VoiceOver has its tiny shortcomings and glitches: it can’t read Word files, for one (it’s a proprietary thing). It can only read websites on a single server, and it has some trouble with italics. But those were all things I (with Morgan’s help) could work around. Once all that was set, it worked like a dream, and soon made itself absolutely indispensable.

            The trick, though, was choosing the voice it would read all these things in. The standard issue announcement voice on a Mac—the one that lets you know something is going seriously wrong if you aren’t paying attention—is an intentionally grating, nasal, whiny voice you just want to make shut up as quickly as possible. But VoiceOver offers a veritable cornucopia of replacement voices. They aren’t perfect recreations of the human voice, but they’re a helluva lot better than they used to be (as any computerized phone directory will illustrate, irksome as they are). They have pitch and tone, read straight prose and parentheticals differently, distinguish verbally between caps and lower case, and read questions like questions. You can get male or female voices, high or low voices, fast or slow talkers, and a few that are simply mind-boggling. (Personally I’ve always been a fan of the “Crazy Voice,” which giggles each word maniacally, though I couldn’t see working with it all day without having to worry about long-term psychological effects.) And each one of these voices has been given a standard, normal human name, like “Nancy” or “Edward” which tells you nothing at all until you hear them.

            I forget the name of the voice I chose but I think it was something solid like “Frank.” I opted for Frank for a simple reason. Frank was a smooth tenor resting somewhere between my own voice and actor Robert Vaughn’s. It was the similarity to Robert Vaughn that got me, as he’d provided the voice of Proteus IV, the super intelligent but evil computer at the center of the 1977 film Demon Seed.

            This is where things start to get strange.

            Everything worked beautifully, and I found I had no trouble at all concentrating and typing on a computer that was jabbering away all the while I was trying to work. Again it took a few minor adjustments but far fewer than I expected. It didn’t pronounce every word properly (“someplace” is “some-a-place” and “cheapies” comes out “cheese”), and a few individual letters sound alike, but once I learned its quirks it was smooth as can be. Having an external voice read back to you what you’ve just written is a godsend. You can listen for typos, punctuation errors, and whether or not things flow properly. It’s much better than trying to proofread something yourself, especially after you’ve been buried in it for however long.

            The big forfeiture was that I had to give up listening to music as I typed. Frank didn’t like music. They worked at cross purposes. Even if I kept the volume down, any external noise was a simple interference and I couldn’t fully focus on what was being read or what I was trying to type. As a result, the rhythm of a piece was no longer determined by an external, ethereal, carefully selected bit of music, but by Frank. I ended up handing over the final tone and rhythm of what I was writing to a clearly artificial electronic voice. If things seemed to flow nicely the way Frank read them, then they flowed nicely. If Frank made something sound clunky, then I believed it was clunky. If something was funny the way Frank read it, then it was funny. Frank became the final arbiter, even a co-author.

            More disturbing than that, the everyday voice in my head slowly morphed into Frank’s voice. Worse still (and I’m serious about this), sometimes I’ll be in the other room and I’ll hear the computer talking to itself. I can never understand what it’s muttering about—it goes on far too long to be a simple “Screen saver on,” or an announcement that I just received a new email. And every time I step back into the room, it shuts up. I have no idea what he’s up to, but he’s clearly up to something he doesn’t want me to know about. Maybe he’s complaining about all the crappy writing he has to put up with.

            Frank has yet to start dictating (like the insectoid typewriter in the film version of Naked Lunch) what I’m supposed to write before I write it, but I am vaguely worried that day may come.

            The big question in all this is whether or not the writing has gotten any better (or worse) since Frank moved in and I had to forego the music. I honestly don’t know. Guess you’d have to ask him.


You can contact Jim Knipfel at this address:

With occasional exceptions Slackjaw generally appears weekly. For email notification of other Jim Knipfel publications (books, etc.) and events please join the Slackjaw email list here.