Glossary

The words of the trade

The ones we use with you when we prepare an event.

Interpreting, AI translation, audience listening and technology: the terms you hear from us, explained in a couple of lines. If one is missing, ask us.

Interpreting

Simultaneous interpreting Also called: Simultaneous
The interpreter translates what the speaker says in real time, with a slight time lag (the décalage). It is used for conferences with many participants.
Décalage Also called: Lag, ear-voice span
The time lag between the original speech and the interpretation. It varies with the language combination.
Relay Also called: Relais, indirect interpreting
Translating between two languages through a third one, the bridge language. If a booth does not cover the speaker's language, it listens to another booth that translates it and interprets from there. For example from Finnish into Slovak via French. It does not cause a noticeable loss of quality.
Pivot
A single language used as relay by all the booths. Pivot is also the name for the interpreter who translates for the audience and for colleagues: when only one booth understands a less widely spoken language, the others listen to its translation and interpret from there.
Retour
Interpreting from your mother tongue into a foreign language. Interpreters usually work into their mother tongue, but some know a second language so well that they also work into it. It is used above all as a bridge between less widely spoken languages and more widely spoken ones.
Cross-translation
A translation from one foreign language into another without going through Italian. It is done by a translator who is a native speaker of the target language and has the source language in their language combination.
Floor Also called: Floor language
The language the speaker is speaking. It can change during the event, even from one talk to the next.

Working languages

Source language
The language of the original speech, which the interpreter translates from.
Target language
The language the speech is translated into.
Active language
The language the interpreter interprets into.
Passive language
The language the interpreter interprets from.
Language combination
An interpreter's set of working languages. AIIC, the International Association of Conference Interpreters, divides them into A language, B language and C language.
A language
The interpreter's main active language, normally their mother tongue.
B language
An active language the interpreter has an excellent command of and translates into in certain situations, even though it is not their mother tongue.
C language
A language the interpreter understands perfectly and translates from. Interpreters do not translate into a C language.

Preparing the event

Kick-off call
The first operational call after confirmation, 30-45 minutes long, in which we align with you on the purpose of the event, languages, format, constraints and context. From here the project is in production.
Context Training
Preparing the AI translation engine on the context of your event before it starts: speakers' names, companies, acronyms, technical terms, taken from your materials and the programme. It is what sets a generic AI translation apart from one made for you.
Floor audio Also called: Floor feed
The main audio signal of the room, from the speakers' microphones through the mixer. It is what the interpreters or the AI engine listen to, and it must be clean: no interval music or announcements.
RSAI
Converso's AI interpreting, in voice and subtitles. The engine translates, and for the whole event one of our linguists monitors the outputs and a technician is on duty at the RSI Hub in Milan.

How the audience listens

BYOD Also called: Bring Your Own Device
The audience uses their own device, usually their smartphone, instead of equipment provided by the organiser. They listen to the translation from the WebApp or the Converso app, with no receivers to hand out and collect.
Token Also called: Event code
In the Converso app, the 6-digit code that gives access to an event's translation. The organiser provides it.
Floor-Through
A feature of the Converso WebApp: you pick your language once and hear the room in that language in your headphones. When the speaker uses it you hear the speaker; when they switch to another, you hear the translation. Subtitles do the same: a transcript of the original, or the translation.
Feed-Through
Former name of the automatic switch between original and translation. It is now part of Floor-Through.
Assisted Listening
The room's original audio, clean, in the headphones of people in their seats, from their own phone. It helps in open, noisy rooms and for anyone who struggles to hear, because of the acoustics or their hearing: everyone sets their own volume. It's Floor-Through used for accessibility.
Voiceover
While you listen to the translation you also hear the original audio in the background, at a lower volume.
Multilingual streaming
A stream, usually on the client's website, in which viewers choose what to listen to: the original or one of the simultaneous interpreting languages.
Floor-In
The web page the room's control room uses to send us the floor, and the video if needed, from a browser: nothing to install. It accepts up to two audio and video streams.
Teleprompter
The web page with real-time subtitles that the control room puts on the screens in the room or on a monitor for the speakers.

Converso platforms

Converso Live
Our video conferencing platform with built-in interpreting: each participant chooses the language they listen to the meeting in.
Virtual booth
In Converso Live, the space for interpreters working on the same language combination: they see and talk to each other, pass terms and hand over turns, with chat, timer and notes reserved for them.
Meeting manager
In Converso Live, the person who moderates the meeting: gives the floor, manages microphones and webcams, changes roles and can remove a participant.
Converso Education
The version of Converso Live for teaching simultaneous interpreting: students work in the interpreter interface, the teacher plays the floor and chooses which student to listen to.
Virtual rooms
Our method for interpreting on platforms that have no interpreter feature: one meeting per language, with translated audio and video crossed between the meetings.

Networked audio and video

SRT Also called: Secure Reliable Transport
An open protocol for transmitting audio and video over the internet. It keeps the stream stable, high-quality and low-latency even when the connection is not perfect.
Caller and Listener
The two roles in an SRT connection. The Listener opens a port and waits; the Caller calls it at its address. To connect, the Listener gives the Caller the address, port and password.
NDI Also called: Network Device Interface
A protocol for transmitting video, and audio, between devices on the same network. At the RSI Hub it brings the room's video to the booths and into video conferences.
Dante
A protocol for carrying multichannel digital audio over an Ethernet network: devices are connected to each other via software instead of dedicated audio cables. At the RSI Hub almost all the audio travels over Dante.

Can't find a word?

Write to us: we'll explain it, and if needed we'll add it here.

Write to us