Typically, while doing Simultaneous Interpreting, the interpreter sits in a booth wearing a pair of headphones and speaks into a microphone. Strictly speaking, “simultaneous” is a misnomer: the interpreter can not start interpreting until he understands the general meaning of the sentence. Depending, for example, on how far apart in the sentence to be interpreted the subject and the verb are located, the interpreter may not be able to utter even a single word until he has heard the entire sentence!
This fact should make it evident how difficult the task of the interpreter really is: he needs to translate the sentence into the target language while simultaneously listening to and comprehending the next sentence. You can experience the difficulty of the task even if you only speak one language: try paraphrasing someone’s speech with a half-sentence delay, making sure you understand the next sentence while paraphrasing the previous one.
One of the key skills of the simultaneous interpreter is decisiveness: there is simply no time to weigh the merits of variant translations or to recall just the right idiom in the target language. Any delay and a few words (and possibly a complete thought) that the speaker uttered could be lost. And since the speaker may be far away, or even in a different room than the interpreter, the loss may be permanent.
During Consecutive Interpreting the speaker stops every 1–5 minutes (usually at the end of every “paragraph” or complete thought) and the interpreter then steps in to render what was said into the target language. A key skill involved in consecutive interpreting is note-taking, since few people can memorize a full paragraph in one hearing without loss of detail. But interpreter’s notes are very different from those of, say, a stenographer, because writing down words in the source language makes the interpreter’s job harder when he has to translate the speech into the target language.
Many professional interpreters develop their own “ideogramic” symbology, which allows them to take down not the words, but the thoughts of the speaker in a sort of language-independent form. Then the interpreter’s output is more idiomatic and less source-language bound.