Your notes aren't a transcript and they shouldn't try to be. They're a map of the speaker's thinking, plus the few things memory can't be trusted with.
New consecutive interpreters tend to make the same mistake: they try to write down everything. By the time the speaker reaches the third sentence, they're still copying the first. Good notes are a map of ideas, not a record of words. Your memory holds the meaning; the page holds the structure, the figures, and the few things you'd regret losing. Here are the principles that experienced interpreters rely on, and a way to build the skill deliberately.
In consecutive interpreting, a speaker talks for a minute or several, then you render the message in the other language. The skill isn't stenography. It's understanding the message well enough to retell it, with notes acting as cues that unlock what you've already understood.
A useful way to think about it: your notes should answer three questions at a glance. What was the speaker's logic? Which facts could I not reconstruct from understanding alone? What's the very next thing they said?
Listen for the point of each sentence and note that. "The budget will be reduced next year because sales fell" becomes a handful of marks, not a sentence in shorthand.
Cause, contrast, consequence, addition. Connectors are what let you retell the speech as an argument rather than a list. Mark them clearly, and use the same marks every time.
Many interpreters note the subject at the left, the verb or action a little to the right and below, and the object below that, stepping down the page as the sentence unfolds. New ideas get a horizontal line across the page. The layout itself shows where you are.
These are the things memory can't regenerate from meaning. Write them in full, in the language you're fastest in, and check them with the speaker if the format allows.
Symbols for recurring concepts (growth, decline, country, time, people, thought, speech) beat abbreviations because they're language-neutral. Keep your set small enough to use without thinking.
If a word's equivalent comes to you while listening, write it in the language you'll speak. It saves a translation step later, when you're under pressure.
Note-taking improves through deliberate repetition, not through reading about it. A routine that works for many students and working interpreters:
Captions don't replace notes. In an in-person consecutive assignment, you'll still do the listening and the noting yourself. But in remote work, or when you practice, a live text feed is a useful second source for the details notes struggle with.
Use it as a safety net and a study aid, never as a crutch. The point of consecutive training is to need the text less each month.
Write the speaker's ideas and the links between them (cause, contrast, consequence), plus anything memory can't regenerate: numbers, names, dates, and lists. Don't try to capture words. Your understanding holds the meaning; the notes hold the structure and the exact details. Unicaption's live captions can serve as a reference for names and figures in remote assignments.
Whichever is faster and clearer for you, but note in the target language whenever the equivalent comes to you naturally, since it saves a step when you deliver. Symbols are language-neutral and work in either direction. Avoid mixing so many systems that you can't read them back under pressure.
It varies with the person and the amount of deliberate practice. Most programs treat it as a skill built over months of regular exercises, not days. Practicing with real speeches, recording yourself, and simplifying your symbol set over time helps more than adding new symbols.
Yes. Run Unicaption on a recorded speech through system or tab audio, interpret it with your own notes first, then compare your version with the live transcript and its side-by-side translation. It supports 60+ languages and stores nothing: audio isn't saved, and transcripts are deleted when the session ends.
Live captions and translation beside any call. 30 free minutes every week.
Start Free Trial →