What is the difference between typing from a written text and transcribing an audio recording? It must be about the same thing. Does this sound familiar?
Is there a difference between typing from a written text and transcribing from an audio recording? Do both kinds of typing involve the same skills? In order to type from a written text, we have to read a word and just type what we see. Transcribing an audio recording requires the ability to hear and listen and understand.
When I think about the differences between typing from a written text and transcribing, it takes me back to 7th and 8th grade, when last period in school, once a week, was Typing. Yes, in an all-girls' private school, it was important to learn the basics, typing in 7th and 8th grade and speedwriting in 9th grade. I never mastered speedwriting. But as it turns out, typing was a good thing to learn. I ended 8th grade with a very marketable skill.
I have memories of sitting down in a room full of manual typewriters, having to identify if I had a pica or an elite (10 points or maybe 12 to anyone who knows what I am talking about), rolling in a sheet of paper, and opening the dreaded workbook, as the teacher recited: f-j-f-j-f-j. What did the teacher's droning recitation teach me? Typing means turning off your mind. Do not try to think ahead or infer what is coming next. The teacher could say anything. She would pick random words, and we would type them. By the end of class, I could be half asleep, thinking about my evening plans and still keep at it: f-j-f-j g-h-g-h.
The skill has stood me in good stead many times. Of course, when given a written text by a client and asked to type it, you really should not turn off your mind; you really should try to type the text correctly even if the client made errors. But sometimes the text is so nonsensical that thinking about what you are typing is just going to trip you up. It is best to just look at one word or one letter at a time, and type what you see.
For example, I recently was given a handwritten manuscript to type for a client who is not a native English speaker, and it was clear that she basically looked up every single word in the dictionary before writing it down, often choosing totally incorrect words. Her grammar skills did not exist; and try as I might, I could not figure out what she intended when she put pen to paper. So turning off my mind, and just looking at the letters helped me to be able to type sentences such as this: "The observation helps to identify process and difficult also locating changing that entered in the process permit to good results aren't prepared and aren't expected."
But what do we do when we transcribe? Not only do we have to listen to the individual words being said and type them correctly, we have to be hyper-aware, listen to the context, and try to make sure that what we are hearing is what the speaker is saying. Generally speaking, my rule of thumb is if it does not make sense, it is probably not correct. That is not to say that every sentence in a transcript must be grammatically correct. In real-time speech, many people start a sentence ... but do not end it. Some people start a sentence in present tense and end it in past tense. There is nothing you can do about people's spoken grammar errors.
Oftentimes, what seem to be grammatical errors are not caused by the speaker per se; the problem is with the ability of a transcriber to hear and/or listen. Maybe the audio was unclear; maybe the speaker's accent is unfamiliar; maybe the speaker is just speaking too quickly; and maybe it is just impossible to tell exactly what was said unless you know what the speaker is talking about in the first place. This problem can often be solved by paying close attention to both the general context and the specific words that are said, and at the same time mentally marking phrases that do not make sense, or adding a timestamp to mark the spot for later review.
I did a bit of research on the topic and could not find any existing definition for the type of listening that is required to properly transcribe. I did, however, find a number of definitions for types of listening. There are many different types of listening, and different people give them different names. Based on most research in the area, the two most fundamental types of listening are:
• Discriminative listening: the most basic type of listening. This is where the difference between sounds is identified. If we cannot differentiate between sounds, we cannot make sense of the meaning that is expressed through such differences. Discriminative listening develops early in life in our native language.
• Comprehensive listening: the next step beyond discriminating between sounds. To comprehend the meaning requires having a lexicon of words at our fingertips along with the rules of grammar and syntax so that we can understand what others are saying.
As someone who has typed many, many papers written by psychology students and students of education, I know that active listening is a term-of-art in psychology, and in fact, it is not really a kind of listening at all. So I have decided to coin the term: "proactive listening" as the type of listening necessary in order to transcribe well.
Proactive listening, I believe, goes a step further, beyond discriminative and comprehensive listening. When transcribing from an audio file, we must pay attention to what we have heard previously and keep it in mind as we continue. By doing so, things that may not be clear at the beginning become clearer later on as we gather more information or hear the same words repeated. Conversely, if we are following the context of what has been said, then something that is unclear later in the proceeding may remind us of something we heard previously, and we can go back and check to see if it fits in the new context as well.
The following are a couple of examples:
• In a transcript that I transcribed in which the audio was slightly unclear, I thought I heard the judge say: "The judgment is final, and that's subject to collateral attack in this court."
A few minutes later, the judge repeated more clearly: "Again, that judgment is final and not subject to attack."
The two sentences contradict each other. So going back and listening again, it became clear that the first sentence actually was: "The judgment is final and not subject to collateral attack in this court."
• In a transcript that I recently edited, the transcriber had typed: "And then you have two blue applications for title."
Half a page down she had typed: "Mr. Smith's application (indiscernible) dated 9/26 states..."
Even if the second instance was not 100% clear, it stands to reason that the sentence should be "Mr. Smith's application for title dated 9/26 states...." Obviously, I could not assume that that was what had been said, but having that in mind allowed me to proactively relisten and hear that, indeed, "application for title" was what had been said.
Again, not every sentence in a transcript will make sense. Sometimes people misspeak. But most of the time people say what they mean and mean what they say. It is our job to make sure what is actually said is reflected properly in the transcripts we produce.
Listening proactively with an active ear and a focused mind allows us to quickly recognize if something makes no sense -- whether it is changing "this debtors" to "these debtors" or realizing that it is not possible someone is talking about a "stocking-house bidder" in the middle of a bankruptcy -- and to fix it. Listening proactively will always result in a better, cleaner, and more accurate transcript that can be certified with a high degree of confidence.
Penina Wolicki is an AAERT Certified Electronic Court Transcriber.