Setup guide
The spoken translation can be played into the loudspeakers of a public address (PA) system or into a hearing loop (induction loop), using a device that plays one language into a compatible line input. It suits a room or zone where everyone needs the same language. This guide covers when it is the right choice, how to connect it, and how to keep the translation from feeding back into SpanVoice.
A PA zone or a loop plays one thing to everyone in it. The translation is one language at a time from one player, so the question to answer first is who in the zone needs it.
Most of the care is in who needs it and in keeping it out of the send.
A PA or a hearing loop carries one programme for everyone in its zone. That suits a room or zone where everyone needs the same translated language, such as a separate room for one language group, an overflow room, or a loop in a room where the people using it have agreed to take the translation.
In a room where the speaker is also heard live, the translated voice arrives a little after the original and the two overlap. For a mixed room, phones with earphones or a headphone system are the better answer. A hearing loop is usually reserved for the original language, so ask the people who rely on it, and whoever looks after it, before you put anything else on it.
A hearing loop (induction loop) serves compatible hearing devices with a telecoil enabled, or dedicated loop receivers. It does not play audibly to the room the way PA loudspeakers do. This guide covers feeding an existing, commissioned loop. It does not cover installing or adjusting the loop itself, and separate loops do not automatically give isolated language zones, because their coverage and interaction need checking by the installer.
The source is the same as for a headphone system: a device that plays one language of the spoken translation. Use the venue’s Audio link, spanvoice.com/a/<venue code>/<language code>, from the venue’s Overview page under For your team or from Links & QR in the desk app, played in a media player such as VLC.
Spoken translation must be on for the venue, under Settings then Spoken translation. Plug the device in, turn on do-not-disturb and set the screen to stay on. The headphone system guide covers the device in more detail.
Connect the player to a compatible line input, using the appropriate lead: a spare line input on the desk or processor that feeds the PA, or the line input on the loop amplifier. Match signal levels as well as connectors. A player’s headphone output is a line-level source, and plugging it into a microphone input will usually overload it.
If you use a DI box, connect its output to the input type its manufacturer specifies. Many DIs give a microphone-level output, which belongs on a microphone input, so the DI’s output level and the input it feeds must agree. A suitable isolating DI can also help with ground-loop hum.
Do not directly join the left and right outputs of a stereo player. Use a suitable stereo-to-mono summing device or cable, or a correctly wired connection from one output channel when the player provides mono audio. Selecting mono on the player changes the audio, not the electrical connection.
The headphone system guide goes through each input type and its lead. For a hearing loop, ask whoever installed or maintains the loop amplifier before connecting anything to it.
This is the step that is easiest to miss. SpanVoice translates what it hears. If the translated voice reaches the send that feeds the desk app, whether from the desk or from a microphone that hears the PA loudspeakers, it will be heard as speech and spoils the result. With one spoken language the extra voice may be left untranslated, but it still disturbs what is translated. With the desk app set to a pair of languages, either as spoken, its words are translated as they arrive and stray words come out in the wrong language.
On the desk, check that the player’s channel is not assigned to the SpanVoice send. Keep the speech microphones away from the loudspeakers, and check that nothing the zone plays can reach them. The mixer guide covers building a speech-only send.
Start with the receiving input turned down. Set the player to a clean output level, then raise the receiving input gradually while checking for overload or distortion. Then walk to the seats. The translation should sit comfortably above the room without distorting, and there should be no ringing or howl when the speaker’s microphone is open.
For a loop, ask the loop maintainer to check the new input without disturbing the system’s commissioned settings. Loop performance involves measured field strength and coverage, not only amplifier volume, so confirm listening quality with a loop receiver and with the people who use the system, and do not judge it from your own ears.
Start a short session, say a sentence into the real microphone, and listen from the audience position. Check that the translation arrives with a delay that is acceptable for your event, that nothing howls, and that the translation does not appear in the captions or the transcript. No howl does not prove the routing is right, because the delayed translation can still be picked up and translated again. Confirm that the player’s channel is excluded from the send to SpanVoice. Then stop the session and start it again, because whether the player resumes by itself depends on the device and the player.
The ten-minute rehearsal goes through the rest.
The same translation can go to several kinds of output. Choose by who needs it.
| Output | Suits | Watch for |
|---|---|---|
| Audience phones and earphones | A mixed room, with any language each person chooses | People bring their own earphones. See the phones guide. |
| Existing headphone system | A room that already hands out receivers | One separate audio feed per language, with compatible transmission equipment. See the headphone system guide and the language channels guide. |
| PA zone | A separate room, overflow room or zone where everyone needs the same language | It overlaps the live speaker where both are heard. Keep it out of the send to SpanVoice. |
| Hearing loop | People using hearing devices with a telecoil, or loop receivers, who agree to take the translation | One programme for the whole loop, usually reserved for the original language. Ask the installer. |
Check the player, then the input, then the send.
Check in order: a session is live and someone is speaking; the player is playing the audio link and its volume is up; the lead and its level match the input (a line source into a line input, or a DI whose output matches the input it feeds); the channel or zone is unmuted and routed to the loudspeakers or the loop.
The microphone is hearing the loudspeakers. Lower the zone level, move the microphone away, or keep the translation out of the zone the speaker is in.
The translated voice is reaching the feed to SpanVoice. Check that the player’s channel is not in the SpanVoice send, and that no open microphone hears the loudspeakers.
The translation arrives after the original. In a room where both are heard, use earphones or a headphone system instead, or use the PA only in a separate zone.
Check the input type and level first, then the gain. A suitable isolating DI, or a ground lift, can help with hum from a ground loop.
Ask whoever maintains the loop. The loop amplifier, the loop itself and the hearing device’s own setting decide what is received, and that is not something SpanVoice can set.
Check the player first: its screen lock, its power saving, its connection. Test background playback on that device before the day.
Still unsure
Tell us the room, the PA and who needs the translation, and we will tell you what to connect.
Talk to usThe free credit lets you try the setup with your own PA. No credit card needed.
A$10 of free credit when you sign up, no credit card needed. Credit never expires — use it this week or next year.