From Janet Zimmermann
Every year or so, New Scientist has a laugh at the expense of television programme subtitlers (Feedback, 23 April).
Live subtitles are generated in one of two ways. Stenographers use a keyboard-like machine, “playing chords” to produce words as they are being said. They have many thousands of key combinations for words and names to remember, and a slip of a finger can result in something that sounds like, but isn’t quite, what was said.
Alternatively, re-speakers repeat what they hear on live TV into speech-recognition software, which generates the subtitles. Although more straightforward than stenography, there is the added complication of listening and speaking at the same time; the slightest stutter or lapse in clarity can result in errors.
Both methods require skill, a lot of practice, preparation, concentration and nerves of steel during transmission.
Advertisement
London, UK
