A few months ago, I bought a dragon. Not the fire-breathing kind, but voice recognition software. Another literary translator I know had been giving it a go, and I was interested for several reasons. Firstly, my hands often tire before my mind does when I’m translating, so I was tempted by the prospect of being more fruitful. Secondly, I hope and plan to continue translating for many years to come, so the longer I can avoid RSI the better. And thirdly, I was intrigued as to whether the change in my physical approach to translating would have any impact on the work I produced. Generally speaking (or as far as I know, at least), translation software tends to be used predominantly in the technical field, but just as athletes look to other disciplines to improve their own performance (one GB Olympic swimmer recently commented that he took up ballet to work on the precise positioning of his arms and legs), why not do that as a translator too? And so the shiny new dragon came into my life…and was then promptly shoved to the back of my drawer for several months. Knowing that it takes time to personalise and optimise it, I was reluctant to devote this time with deadlines approaching from all sides. But a few weeks ago, as I was about to start the first draft of a novel, the time was right. Since then, the dragon and I have been getting to know each other,  and I felt compelled to share a review. Disclaimer: I haven’t used voice recognition software for translation before, and am certainly no expert in comparing the various kinds, so these are my personal, cursory and unprofessional reflections.

First, the dragon takes you through a set-up process, in which you can familiarise it with your voice, as well as direct it to read documents resembling those you will most commonly dictate. There are a vast number of tools and features, and I can tell I’m not yet pushing my dragon to its full potential. But it’s performing wonderfully for what I need it to do. The last time I used any kind of recognition software was during an internship at a legal firm in the 1990s, and the technology has definitely come on a long way since then. Although there are certainly times when it gets the word wrong, most commonly with unfamiliar terms such as foreign names, place names and so on, I’ve found it to be very responsive. And if you put the time into correcting it, it quickly learns and provides the right response in subsequent dictations. You can also add specific words to its internal vocabulary. Having given it my translation of a previous novel by the same author to read as homework, I was pleased to see it picked up on some character and place names right away in the new novel.

I admit that the dragon and I are probably still in our honeymoon period, but I can already tell that it’s helping me to produce first drafts more quickly. And it’s not a case of sacrificing quality for quantity; instead, it’s enabling my printed output to keep up with my mind. I’m pretty sure that even the speediest of touch typists (of which I’m certainly not one) can’t type as fast as they can think. Another feature which could be handy for translators (although I haven’t tried it yet) is the playback option, whereby you can listen to a recording of your dictation.

But the most interesting thing is that I feel like this method is improving the quality of my first drafts, and that’s much more beneficial than any increase in speed. I believe this is because I’m reading my text aloud in the very first stages of the translation process. Some translators, depending on their own personal approach to first drafts, might not like the shift, but it seems to be working for me. I’m not using it for everything, as it suits some texts more than others, but I like the freedom of being able to switch between dictating and typing according to how I feel. Given that the dragon responds better to fluid, naturally read sentences than it does broken ones, I find that I’m being much more thoughtful and sensitive to the cadence of my words in the first draft stage, which means the subsequent editing stages can focus more on the fine-tuning. It also feels like this new method is boosting my creativity, refreshing my mind and the way I look at a text. And that can only be a good thing.

Update: A couple of points to add. Firstly, just to confirm that I don’t have any business connections of any kind with Nuance, the software company. And secondly, I’ve heard that the dragon can be a bit of a prude when it comes to swear words, so that may be something for the developers to work on!

One thought on “In which I train a dragon…

  1. Pingback: Weekly favorites (Aug 6-12) | Adventures in Freelance Translation

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s