Editing Speech Recognition Transcripts--Multitaskers Beware!
Over the past few years I've had the opportunity to speak to MT groups across the country and write professional journal articles on the realities of speech recognition editing. Since the vast majority of my live audiences are women, I like to break the ice by asking the question, "What's the one thing that women generally do far better than men?" (Of course, there's always someone in the crowd who hollers out, "Everything!") True as that may be, the answer I'm looking for is, of course, multitasking. It's no secret that most women's brains are hard-wired for multitasking while most men have what can only be described as one-track minds. Most female MTs I know can transcribe with "half their brains tied behind their backs" while using the other half to create a shopping list, balance the checkbook, and mentally sort through any number of other personal or professional issues. This ability to multitask seemingly effortlessly is an attribute I envy, quite frankly, and greatly admire in my wife and all other women.
When it comes to speech recognition (SR) editing, however, it's been my experience that mental multitasking is more of a burden than a blessing, especially for experienced MTs transitioning from straight transcription to SR editing. The reality is that what I call the ear/eye/brain/hand (EEBH) dynamic of editing is markedly different than that of traditional transcription. I can't describe the difference to you in scientific terms, but take my word for it, the contrast is very real.
There's a certain familiar process that takes place when one is listening to dictation, typing the words on the keyboard, and watching the text appear on PC monitor. For many experienced MTs of the XX chromosomal persuasion, this process doesn't seem to require a huge amount of conscious effort, thus freeing up the excess mental capacity for other tasks. It's like having some extra RAM on board, you might say...
The process of editing SR drafts alters the EEBH interaction significantly, however. In most cases, the editor is listening to the original dictation, as with straight transcription, but after that things start to get synaptically complicated. The feedback loop alters because the editor's hands aren't creating the text on the screen in real time. As a result, the brain often gets conflicting signals from the eyes and the ears; the brain thinks the eyes are seeing what the ears are hearing, when in fact that may not be true. Exacerbating the problem is the fact that many editors increase the playback speed of the dictation in order to churn out more lines. I've experienced this phenomenon myself, have personally observed it in numerous MTs under my supervision, and have heard the same story from many others in the industry.
The net effect of all this is that many experienced MTs will find it takes considerably more mental concentration to accurately edit SR transcripts than it does to transcribe from scratch. This, by the way, is one reason I'm quick to point out that editing isn't an "easier" version of transcription. Certainly it's true that SR editing generally requires fewer keystrokes per line (although there are a number of variables to that factor). But it's also true that because of the added mental strain, many MTs may find that SR editing is "harder" than traditional transcribing. Disclaimer: These terms are subjective, and your mileage may vary.
Bottom line, for many MTs transitioning from straight transcription to SR editing, dialing back on the mental multitasking may result in a better editing experience and help eliminate some of frustration inherent in making the switch.