Against on-screen subtitles
Reading is not the same as listening. Live subtitles work, and they have an attention cost that shows the moment an event runs past twenty minutes.
It is the most frequent comparison, because putting automatic subtitles on a screen is easy and cheap. The question is not whether they are understood, but what the audience is doing while understanding them.
Where each one stands
Where subtitles win
They are seen without earphones, they work in a noisy room, and they serve people who cannot hear. And one screen is shared by everybody without anyone having to do anything.
Where the voice wins
It is heard while looking at whoever is speaking. With subtitles the audience watches the screen and misses the person, the slides, and whatever is being pointed at.
The tiredness
Reading subtitles for an hour is far more tiring than listening for an hour. At a long event half the room gives up on the screen and follows in fragments.
You do not have to choose
Here both run at once: translated voice on each person's phone and the text on that same phone, for whoever prefers to read or does not hear well.
When subtitles are enough
At a short event, in a room where nobody is going to put earphones in, or when the main audience is deaf. And if the content is very visual — figures, charts, demonstrations — reading beside what is being shown gets in the way less than it does in a talk.
Questions people ask
- Can I have both?
- Yes, and it is the normal case: each listener sees the text on their phone while hearing the voice. There is no choosing on behalf of the whole room.
- Do subtitles arrive before the voice?
- The text appears before the translated audio, because generating the voice adds its own time. The gap is under a second.
- Can subtitles go on a big screen?
- What we give is live text per room and per language. Getting it onto a screen in the room is for whoever builds the event, and with a single screen you have to pick one language.