We’ve seen four interfaces with voice output through speakers so far.
- The message centre in the New Darwin hotel room, which repeated the onscreen text
- The MemDoubler, which provided most information to Johnny through voice alone
- The bathroom tap in the Beijing hotel which told Johnny the temperature of the water
- The Newark airport security system
Later, in the brain hacking scene, we’ll hear two more sentences spoken.
Completionists: There’s also extensive use of voice output during a cyberspace search sequence, but there Johnny is wearing a headset so he is the only one who can hear it. That is sufficiently different to be left out of this discussion.
Voice is public
Sonic output in general and voice in particular have the advantage of being omnidirectional, so the user does not need to pay visual attention to the device, and, depending on volume and ambient noise, can be understood at much greater distances than a screen can be read. These same qualities are not so desirable if the user would prefer to keep the message or information private. We can’t tell whether these systems can detect the presence or absence of people, but the hotel message centre only spoke when Johnny was alone. Later in the film we will see two medical systems that don’t talk at all. This is most likely deliberate because few patients would appreciate their symptoms being broadcast to all and sundry.
Unless you’re the only one in the room
The bathroom tap is interesting because the temperature message was in English. This is a Beijing hotel, and the scientists who booked the suite are Vietnamese, so why? It’s not because we the audience need to know this particular detail. But we do have one clue: Johnny cursed rather loudly once he was inside the bathroom. I suggest that there is a hotel computer monitoring the languages being used by guests within the room and adjusting voice outputs to match. Current day word processors, web browsers, and search engines can recognise the language of typed text input and load the matching spellcheck dictionaries, so it’s a fair bet that by 2021 our computers will be able to do the same for speech.