How to connect a mixer to your computer.

Connect your mixer to a computer in five steps

Most of the work is on the desk. Save it in a scene and it is there next time.

1. 1

   Pick a spare output on the desk and make it a speech-only mix

   SpanVoice needs the words and nothing else. Do not use the main left/right output: it carries the music, the playback and every change the operator makes during the event. Use an output that is doing no other job in the room:

   • A spare aux send, set post-fade so it follows the channel faders: when the operator pulls a microphone down, SpanVoice hears less of it too. Whether a muted channel also leaves the send depends on the desk; most mute post-fade sends, and the test in step four confirms it on yours.
   • A matrix output, fed only from the speech channels or a speech group.
   • A group or subgroup that holds only the speech microphones, taken from its direct output.

   On that send, bring up only the microphones people speak into: lectern, handheld, headset, lapel. Leave music, playback, click and effects returns at zero. Take the send pre-effects so no reverb or delay reaches it. Mono is right; one output is all you need. Put a strip of tape on it that says SpanVoice so nobody repurposes it on the day.

   Speech mics lectern, handheld, lapel

   XLR

   Sound desk aux or matrix post-fade, pre-FX

   one output

   Aux out mono, line level

   If the desk recalls a saved scene at the start of the day, save the scene with the SpanVoice send in it. A recall that forgets the send is the most common reason the words stop before the event has begun.

2. 2

   Connect the output to the computer

   A mixer puts out line level. The computer needs to take it in at line level, not through a microphone socket. There are three ways, in order of how well they work; the table under the steps compares them.

   1. A small USB audio interface between the desk and the computer. Run a balanced lead from the aux output to a line input on the interface: XLR or 6.35 mm TRS at the desk end, and a 6.35 mm TRS plug at the interface end, which on a combined socket is the centre jack (the XLR side is often microphone-only). Make sure the input is on line, not instrument, and turn its gain most of the way down. The interface appears in Windows as a recording device. This is the one to buy if you are buying anything.
   2. A mixer with built-in USB audio, digital or analogue. The mixer is then the sound card and the computer sees it as a recording device. What reaches the USB channels varies: many small analogue mixers send only the main mix over USB, while digital desks let you route any bus to any USB channel. Check the routing page or the manual; if the aux send cannot be routed to USB, use an interface on the aux output instead.
   3. The computer’s own 3.5 mm socket, only if it is a true line input. Most laptop sockets are headphone-only or a headset connection with a microphone contact, and no cable fixes that. Where a labelled line-in exists, a 3.5 mm lead from the aux output works for a test; for anything you rely on, use an interface.

   

   Sound desk aux out, line level

   XLR / TRS

   USB interface line input, gain low

   USB

   Windows computer the desk app

   A balanced cable (XLR or TRS, three conductors) keeps hum out over a long run between the desk and the computer. An unbalanced cable is fine for a metre or two on the desk.

3. 3

   Set a clean speech level at the mixer

   Open the desk app on the computer, choose the interface in the Input panel, and watch the meter. It runs before you start streaming, so nothing is sent while you do this. Have somebody talk at the microphone at the level they will use on the day, the loudest speaker you expect.

   • Adjust the send or output level on the desk, then the interface gain, and leave the Windows recording level alone unless nothing else is available.
   • Aim for the meter moving clearly into the green on every sentence, with the red segments never lighting and room to spare for a louder moment. The reading beside it is the loudest recent level in decibels. Check the desk and interface for clipping as well: a healthy-looking meter in the app does not prove the signal was clean earlier. The audio level guide explains what the meter shows.
   • Ask them to stop talking. The meter should fall almost to nothing. If it keeps moving, the send may be carrying the room, a music channel or an open microphone, or noise. Fix that on the desk, not in the app.

   Save the desk scene with this send in it, and check the meter for a few seconds before each event. The send is separate from the main mix, so it will not drift on its own, but a recalled scene or a new microphone can change it.

   Louder is not better. A feed that overloads loses words; a steady, clean feed with room to spare is what SpanVoice is built for.

4. 4

   Rehearse from the desk app

   Press Start streaming and read the words on your own phone from the venue’s audience link. Say a sentence with a name in it and a number in it, and watch both arrive. Then mute the microphone on the desk and watch the meter in the app: if it drops, mute follows the send on your desk; if it keeps moving, mute does not reach the send and the operator needs to pull the fader down as well. If nobody is around to talk, switch the Input panel to Test file and play a recording of a sentence or two instead; it is sent exactly as a microphone would be.

   There is no way to listen to the live input on the computer itself. To hear what SpanVoice hears, solo the SpanVoice send on the desk’s headphone output.

5. 5

   On the day, check the meter and leave the rest

   The desk output is now the one thing nobody needs to touch. Start streaming when the event starts, glance at the meter now and then, and press Pause audio during music or a break if open microphones are picking up the room. When a new microphone is added to the room mix, add it to the SpanVoice send as well, or that person’s words never appear.