Remote simultaneous interpretation, often shortened to RSI, is simultaneous interpretation where the interpreters are not in the room. They listen to the speaker over the internet, interpret in real time from wherever they are, and their voice is sent back to the audience through a digital platform. The work itself is the same as in a booth. What changes is where the interpreters sit and how the sound travels.
How remote simultaneous interpretation works
In a classic setup, interpreters sit in a soundproof booth at the back of the venue. They hear the speaker through a headset, speak into a microphone, and the audience listens on receivers handed out at the door. RSI keeps the human interpreter but replaces the booth and the cables with software.
The source audio is captured at the venue or from an online meeting, usually from the sound desk or the speaker's microphone, and streamed to the platform.
The interpreters connect remotely, from a home studio or from a hub equipped for the purpose. They hear the speaker, often see a video feed, and interpret into their language as the speech unfolds.
The interpreted audio is distributed to the audience. Participants pick their language channel in an app, a web page or a video conferencing tool, and listen on their own device or on venue receivers connected to the platform.
As in a booth, interpreters generally work in pairs and take turns, because simultaneous interpretation demands intense concentration. The platform gives them a way to hand over the microphone and to communicate with each other without the audience hearing it.
Why organizers choose RSI
RSI became common because it solves real problems of on site interpretation.
No travel for interpreters. You can work with qualified interpreters wherever they live, which widens the pool, especially for less common language pairs.
Less hardware in the room. Without a physical booth, the venue needs less space and less installation. That matters in rooms where a booth simply does not fit.
Online and hybrid audiences. Because the interpretation already lives on a platform, remote participants can follow it as easily as those in the room.
Human judgment stays in the loop. A professional interpreter still makes the choices: tone, register, cultural nuance, what to stress and what to leave out. For sensitive negotiations or highly technical debates, that judgment is the whole point.
The constraints you should plan for
Moving interpreters out of the room does not remove the complexity. It moves it somewhere else.
Connection. Every link in the chain depends on the network: the feed from the venue, the interpreter's connection, and the audience's access to the platform. A drop on any side means a gap in interpretation. Serious setups plan a backup connection at the venue and ask interpreters to work on a wired line.
Sound quality. Interpreters can only translate what they hear clearly. A weak microphone, a speaker who walks away from the lectern or a noisy room makes their job much harder, and the result is less accurate. Remote interpreters cannot lean over and ask a technician to fix it. Someone at the venue has to watch the audio closely.
Coordination. Interpreters need the programme, the slides and the names of speakers in advance. They need a technician or a moderator on the platform side to manage channels and handovers. Booking is still done well ahead, often weeks before the event, and changes to the schedule have to be passed on to every team.
Cost per language. RSI saves on booths and travel, but each language still needs its own interpreters. Covering several languages means several teams, each booked and paid for the full session.
How RSI differs from live AI translation
Live AI translation, the kind of service Glot provides, answers the same need with a different method. There is no interpreter, remote or otherwise. Software listens to the speaker, transcribes what is said, translates it sentence by sentence and delivers the result as captions, and in some offers as a synthesized voice.
The differences are practical.
Who does the translating. In RSI, a person interprets. In AI translation, a machine translates. A human interpreter can adapt to an ambiguous phrase, a joke or a politically loaded word in a way software still cannot fully match. On the other hand, the machine does not tire as the day goes on.
How languages scale. With RSI, each new language means a new team. With an AI service, adding a language usually means selecting it. Glot, for example, covers more than 60 languages, all included in the price, so a room can serve a dozen nationalities without changing the setup.
What the audience receives. RSI delivers an audio channel. AI translation typically delivers text first: live captions in each person's language on their phone, and often captions on the room screen. Some services add a translated voice for those who prefer to listen.
Preparation. RSI needs interpreters to be briefed and scheduled. An AI service can be set up in minutes, provided the audio is clean. Participants usually just scan a QR code and follow in their phone's browser.
Delay. Both approaches run slightly behind the speaker. An interpreter waits for enough of a sentence to understand it. A translation engine does the same thing for the same reason. With Glot, the delay is around two seconds.
Both approaches share one weakness: they depend on good sound. A clean feed from the sound desk or a dedicated microphone helps an interpreter just as much as it helps software.
Which one fits your event
RSI makes sense when the stakes call for human judgment on every sentence: diplomatic meetings, legal proceedings, board discussions where a single word can carry weight. It also fits when you need an interpreted audio channel above all, and the budget and lead time are there to book teams per language.
Live AI translation fits the many events where the goal is to let everyone follow what is said, in their own language, without a long setup: conferences, internal meetings, training sessions, product launches, events with an audience from many countries. It is also a way to cover languages you could never staff with interpreters.
The two are not always rivals. Some organizers use interpreters for the main language pair and AI captions for everyone else. The useful question is not which technology is better in general, but what this event, this audience and this budget actually need. If you want to compare with your own numbers, the pricing page shows what a room costs per hour or per day.
Want live captions and translated voice in more than 60 languages, set up in minutes?
See how Glot works