Audio X
User Manual
Version 1.4.0 · 4 November 2026
Copyright © 2026 Justin L Worley · GNU Affero General Public License v3
audiox.info · contact@audiox.info
audiox.info · contact@audiox.info
Audio X is a Windows program for playing audio in live venues. It runs a cue list for a show, an automatic radio station with idents, a soundboard, a talkback microphone, a multitrack recorder, a real-time analyser and sound level meter, and it controls other equipment (Yamaha TF consoles, MIDI devices and Q-SYS) from the same cue list. You can watch and control it from phones, tablets and other PCs on the same network.
This manual covers version 1.4.0.
Copyright (C) 2026 Justin L Worley. Audio X is free software under the GNU Affero General Public License, version 3. See Licence and Copyright. Contact: contact@audiox.info.
Start here
Preparing audio
Live operation
Connecting to other things
Reference
Run AudioX-Setup-1.4.0.exe and follow the steps. A portable zip (AudioX-1.4.0-win-x64.zip) is also supplied: unzip it anywhere and run AudioX.exe. Your settings are kept in %AppData%\AudioX and your shows are .axshow files wherever you save them.
The first time Audio X listens on the network (for the remote control or the stage display), Windows asks whether to allow it. Choose Private networks.
Audio X needs an audio device before it can play. Open Settings > Audio device.
In Preferences you can tick Open the audio device automatically so Audio X opens it every time it starts.
For a Yamaha TF console connected by USB, press Open Yamaha TF5 (USB): it opens the driver with all 34 channels. See Input Monitor and Recorder. For Dante, see Troubleshooting.
Every cue has an Output drop-down. The default is Ch 1 + 2 (stereo). A mono file plays to both channels of a pair. You can send a cue to another pair, a single channel, or all outputs.
| Item | What it does |
|---|---|
| + Add Cue(s) | Adds audio files to the cue list, one cue per file. |
| Edit Cue | Opens the Cue Editor for the selected cue. |
| Process List | Converts every cue to MP3, trims silence and normalises the loudness. See Process List and Normalise. |
| Show, Cues, Live, Audio, Network, Settings | The six menus. See the Menu Reference. |
| Clock | The show time (the PC clock, corrected by the NTP time server when it is on). Underneath: the SMPTE timecode being sent or the time source. Click it for the timecode settings. |
| Stage / Open | Choose a screen and press Open to show the stage display on it. |
| NTP | Tick to keep the clock in step with a network time server. |
TRT is the total running time of all cues. REMAINING is what is playing now plus every cue from the standby cue on. Looping cues count once.
| Button | Key | What it does |
|---|---|---|
| GO | Space | Plays the standby cue and moves standby to the next cue. |
| STOP | Fades out everything that is playing. | |
| PAUSE | Pauses; press again to resume. | |
| PANIC | Esc | Stops cues, the tone generator and the soundboard at once. The radio and a recording keep running. |
All keys can be changed in Settings > Key mapping. F11 switches full screen on and off.
| Column | Meaning |
|---|---|
| # and Name | The cue number and name. Double-click to edit. An orange arrow marks a cue that is playing; a blue bar marks the standby cue. |
| Output | Where the cue plays. |
| Gain dB | Level of the cue. Normalise and Process List set it for you. |
| Fade in / out ms | Fade times. |
| Loop | Repeats the cue until stopped. |
| Then | None: nothing. AutoContinue: when this cue starts, the next cue starts too. AutoFollow: the next cue starts when this one finishes. |
| Actions | How many show-control actions the cue fires. |
| Duck | Bed (music that the radio ducker turns down), Trigger (an announcement that turns beds down), or None. |
| Progress, Time | How far a playing cue has got and its time left. |
5, 1:30, 1:05:00, 90s, 2.5m, 1h) and press Set. Start starts, pauses and resumes; Reset clears. It flashes orange in the last 10 seconds and red at zero, and shows on the stage display and the web page.The ten pads at the bottom are the first ten pads of the soundboard bank chosen in the little Bank drop-down. Click a pad to play it, right-click to edit it, click an empty pad to assign a sound. See Soundboard.
Shows the audio device (rate, buffer, channels, latency, CPU), the link state of the Yamaha console, Q-SYS and MIDI, and the live sound level when the analyser is running (it says OVER LIMIT when the limit is exceeded). The version is at the bottom right.
A show is a cue list saved as a .axshow file. A cue plays one audio file (or a group of files), or does something to other equipment without playing audio.
Each column of the cue list is a setting: output, gain, fades, loop, then, duck. For trims and waveform work use the Cue Editor.
Cues > Cue actions opens the actions of the selected cue.
Choose an action type (Yamaha scene, fader or mute, raw RCP line, MIDI program / control / note / SysEx, Q-SYS control or snapshot), fill in its values and set a delay in milliseconds. All actions of a cue start when the cue fires. See Show Control.
.axshow file. Audio files are referenced, not copied: keep them where Audio X can find them.Select a cue and press Edit Cue. The editor never changes your audio file: it stores trim, fades and gain in the cue.
The waveform shows the whole file. Drag the start and end markers to trim, or type the times. Anything before the start marker or after the end marker is not played.
Audio X leaves a tiny margin (10 ms before the first sound, 50 ms after the last) so nothing is clipped.
Set fade in and fade out in milliseconds and choose the fade curve. Gain is the cue's level in dB; Normalise sets it so the trimmed part reaches a target loudness.
Preview plays the cue through the audio engine with the trims and fades applied.
For real audio editing (cutting, effects, mixing) press Edit audio in OpenDaw. Audio X opens the file in OpenDaw, a separate free program, and reloads the waveform when you come back. See Audio Tools.
The Process List button (top bar) gets a whole cue list ready for a show. It runs three steps on every cue, in this order, so the loudness is measured on the finished, trimmed file.
| Step | What it does |
|---|---|
| 1 Make sure all audio is MP3 | Any file that is not an MP3 is converted to MP3 at 48000 Hz, 256 kbps, stereo (you can change the values). Files that are already MP3 are left alone. A file used by two cues is converted once. |
| 2 Auto trim | Moves the start and end of each cue to the first and last sound, using the silence threshold (default -50 dBFS) and the same margins as the Cue Editor. |
| 3 Normalise | Sets each cue's gain so the trimmed cue plays at -18 LUFS, with the true peak kept below -1 dBTP (the same standard values as Normalise all cues). |
Tick or untick each step, change the values, and choose All cues or Selected cue only. Then press Process.
The table shows what happened to each cue. In this example the cue Act 1 Opening had 1.2 s of silence at the start and 1.8 s at the end, which were trimmed, and its quiet recording was lifted by 10.9 dB. A cue with no audio (an action cue) is skipped.
Your original files are never changed. Converted copies are saved next to the originals with the extension .mp3; the cue list is pointed at the copies. Trims and gains are stored in the cues. Undo puts every cue back as it was (the converted MP3 files stay on disk).
Process List needs FFmpeg to convert (see Media Converter). Trimming and normalising need no FFmpeg. Several files convert at the same time; the number is set in Hardware Acceleration.
Cues > Normalise all cues measures loudness (ITU-R BS.1770, integrated LUFS and true peak) without changing any files.
Use Normalise when you only want levels. Use Process List when you want conversion, trimming and levels together.
Audio > Media converter converts audio and video files to other formats. It scales pictures up or down, changes codecs and containers, and uses the graphics card to encode video when it can.
Several files are converted at the same time (see Hardware Acceleration). The summary line at the bottom gives the total time.
| Group | Choices |
|---|---|
| Audio | WAV (16, 24 or 32-bit float), FLAC (16 or 24-bit), MP3, AAC (m4a), Ogg Vorbis, Opus |
| Video | MP4 H.264 + AAC, MP4 H.265 (HEVC) + AAC, MKV H.264 + AAC, MKV H.265 + AAC, MKV AV1 + Opus, WebM VP9 + Opus, MOV ProRes 422 HQ + PCM, MOV H.264 + AAC |
| Re-wrap | MP4 copy video + AAC: the picture is not re-encoded, so there is no loss and it is fast |
A video source converted to an audio format keeps only the audio (the first audio track).
Tick Use the graphics card (GPU) where possible (it appears for video formats). H.264, H.265 and AV1 are encoded on the GPU when this PC has a card that can do it; the line next to the tick shows what was found, and the Encoder column shows what each file used (for example hevc_nvenc). If the card cannot do a job, the CPU does it. Hardware settings opens Hardware Acceleration.
The converter uses FFmpeg. Audio X looks for it in the path you set, then the Audio X folder, then the system PATH, then the WinGet links folder. If it is not found, the window says so; install FFmpeg (for example winget install Gyan.FFmpeg) or press Set ffmpeg and choose ffmpeg.exe.
Settings > Hardware acceleration controls how Audio X uses the PC's graphics card and processor for batch work.
The window lists the graphics cards in the PC and the GPU video encoders that really work. Audio X test-runs each one on a tiny clip (NVIDIA NVENC, Intel Quick Sync, AMD AMF; H.264, H.265 and AV1) instead of trusting a list, so only encoders that start are offered. Look again repeats the check, for example after a driver update.
The Media Converter then encodes H.264, H.265 and AV1 on the card. In Audio X's own tests a 20-second 1080p clip encoded to H.265 about four times faster on an NVIDIA RTX 3090 than on the CPU. Your speed-up depends on the card and the footage. Pictures are scaled on the CPU.
Files processed at the same time (0 = automatic) applies to the Media Converter and to Process List. Higher numbers use more processor cores.
MP3 conversion, trimming and loudness analysis are processor work with no graphics-card version. For these jobs the speed comes from processing several files in parallel. Audio playback, mixing and the microphone processing also run on the CPU, in Audio X's real-time audio engine.
Live > Radio and idents runs an automatic radio station: a playlist of songs with idents, promos and announcements mixed in between, with no gaps.
Idents are short sounds that play between songs.
| Column | Meaning |
|---|---|
| On | Switch an ident on or off. |
| Name, Files | The name and how many audio files it has. Files rotate randomly, never the same one twice in a row. Press Files to choose them. |
| Trigger | EverySongs: after every N songs. ClockTime: at a time of day. |
| Every N songs / Clock time | The N, or the time (for example 00:00 for the top of the hour; it uses the NTP-corrected clock and plays once). |
| Output, Gain dB | Where the ident plays and its level. |
When TALK is held (button, phone page or key) or an announcement cue plays, the music is turned down so the voice is clear, then comes back.
Spotify needs a Premium account and a free Spotify developer app.
http://127.0.0.1:8888/callback).spotify:playlist:...) into the window and press Connect. Your browser opens so you can sign in there.Spotify cannot take an ident between songs, so idents play over the music (which is ducked), one after the other. Spotify's terms restrict public and commercial performance: check your venue licence, or use the local playlist.
Live > Soundboard has 4 banks (A to D) of 16 pads. A pad starts in about 25 ms because its sound is already loaded.
Drop an audio file on a pad to assign it. Press Edit pads (or right-click a pad) to set:
1-2, 3, 1,2,5-6 or allThe corner of a pad shows HOLD or LOOP for those modes.
While a pad that has duck ticked plays, the cues, radio and loopback audio on the same outputs is turned down, then recovers. The settings at the bottom of the window are Depth (default 12 dB), Attack (30 ms), Hold (250 ms) and Release (600 ms). Other outputs are not touched. A mono pad plays on every selected output; a stereo pad uses the first two.
Audio > Channels and microphones sets the two microphones Audio X can use. They are different devices, so each has its own choice.
This is the microphone used for announcements. While TALK is held, it is heard on the outputs chosen in the Radio window and the radio music is turned down. The microphone can be:
It works whether or not the radio, a cue or the soundboard is playing.
The announcement microphone is processed all the time, so it is already settled when you press TALK. The chain is:
| Stage | Purpose |
|---|---|
| Low-cut (default 100 Hz) | Removes rumble and the low notes that start feedback. |
| Feedback suppression | Finds a tone that keeps ringing and cuts a narrow notch there (up to 6, learned in the room). Voices are not mistaken for feedback. Forget learned feedback notches clears them. |
| Noise gate | Turns the room down between words. By default it follows the room's noise by itself; untick Gate follows the room noise to use the manual dB value. |
| Auto gain | Brings every voice to a steady, loud level: target level (default -20 dB), up to a maximum boost (default 36 dB). It only moves while someone is speaking, so the room noise is not pumped up. |
| Compressor | Evens out the voice. |
| Limiter | Nothing louder than the ceiling (default -1 dB). |
Every stage can be switched off, and Process the announcement mic switches the whole chain off. Press Apply microphone settings to save and use the changes.
The Radio window and the phone page show a live level bar and a sentence about the microphone, for example the levels and gain, or Input 3 is not open on the audio device. If TALK does not work, read that sentence first.
A phone or tablet microphone needs the secure page (https) because browsers only allow microphones there. See Remote Control and the Phone App.
The microphone for the analyser and sound level meter. Choose a separate capture device (for example a measurement microphone) or a channel of the main device, and its input channel. The analyser is restarted when you press Apply if the Monitor is running.
Live > Monitor is a real-time spectrum analyser (RTA) and a sound level (SPL) meter.
The microphone and channel are chosen in Audio > Channels and microphones (the boxes at the top of the Monitor window show the choice). Press Start to begin. The input level bar and the CLIP flag show the raw microphone level: keep it out of the red.
Put a calibrator (94 dB at 1 kHz is common) on the microphone, enter its level and press Calibrate now. Until then the window says UNCALIBRATED - levels are relative. The offset is saved. Mic file loads a frequency response file (for example the UMIK-1 calibration text) to correct the spectrum.
1/3, 1/6, 1/12, 1/24 or 1/48 octave bands from 20 Hz to 20 kHz, A / C / Z weighting, averaging (Off, Fast, Slow, Long), Peak hold, and adjustable top level and span. Move the cursor over the graph to read a value.
The analyser is shown (view only) on the RTA tab of the phone page and as a tile on the confidence monitor.
This is not a certified sound level meter. Accuracy depends on the microphone and calibration. Use a measurement microphone for real work (a laptop or webcam microphone has automatic gain) and a certified meter where a legal limit applies.
Both need an audio device with input channels open (for example ASIO > Dante Virtual Soundcard or the Yamaha Steinberg USB driver, with Input channels set in Settings > Audio device).
Live > Input monitor shows level bars (-60 to 0 dBFS, peak hold, clip light) for the inputs you choose.
Live > Multitrack recorder records up to 48 tracks.
Each track is patched to an input channel of the main device with the track's Input drop-down. Patch 1:1 patches track n to input n; Clear patch empties the table. Meters for every patched track are live as soon as the window opens.
For each track you have R (record enable), I (input monitoring), S (solo), M (mute), Ø (polarity), trim, fader, pan, a colour and a name. Click a track to open its strip: bus, aux sends, high-pass, 4-band EQ, gate, compressor and limiter. These shape what you hear in the monitor mix only: the recorded files are always the raw input (an option can print trim and polarity into them).
Get track info from Yamaha TF5 reads the channel names and colours from the console (its address is set in Settings > Show control). Names become the file names. If you press Record from TF5 (USB) and the address is known, names are read automatically.
SessionName_yyyy-MM-dd_HHmmss, writes all the files there and adds session.json when you stop.A recording keeps going if you close the window, PANIC does not touch it, and Audio X will not close while a recording is running.
| Choice | Notes |
|---|---|
| WAV | Most compatible. A file that passes 4 GB converts itself to RF64. |
| BWF | Broadcast Wave with time reference, iXML (project, scene, take, note, track names) and markers. |
| RF64 / BW64 | 64-bit WAV for very long recordings. |
| 24-bit PCM or 32-bit float | 32-bit float is a bit-exact copy of the input and does not clip above 0 dBFS. |
| Mono files or one polyphonic file | Files are named NN_TrackName.wav. |
The sample rate is the open device's rate. File headers are refreshed every two seconds, so a crash or power cut still leaves playable files. If the disk ever stalls for longer than the buffer (about 5 seconds at 48 kHz), whole blocks are dropped for all tracks together so they stay in sync, and the window shows DROPOUTS.
Remux converts a finished session to another container, bit depth or file layout without touching the originals. Open in OpenDaw opens the session for editing and mixing (see Audio Tools).
Install the Yamaha Steinberg USB Driver once, then connect the console by USB. Press Record from TF5 (USB): Audio X opens the driver at 48 kHz with all 34 inputs and patches USB n to track n. Only one program can use an ASIO driver at a time, so close any other program that has the console open.
A cue can do things to other equipment when it fires. The equipment is set up once in Settings > Show control, then each cue gets actions.
get MIXER:Current/InCh/Fader/Level 0 0).The bottom status bar shows the state of all three. A device that is offline is skipped for five seconds after a failure so a dead console cannot delay every cue.
Select a cue and open Cues > Cue actions, or add a cue with no audio using Cues > Add action cue. Every action has a delay in milliseconds; all actions of a cue start when the cue fires, each on its own timeline.
| Action | What it does |
|---|---|
| Yamaha scene | Recalls scene A or B, number 1-100. |
| Yamaha fader | Sets Input 1-40, Mix 1-20, Matrix 1-4 or the Stereo master to a level from -138 to +10 dB, or -inf. |
| Yamaha mute / unmute | Mutes or unmutes a channel or bus. |
| Yamaha raw | Sends any RCP line exactly as typed. |
| MIDI | Program change (with optional bank), control change, note (with automatic note-off) or SysEx. |
| Q-SYS control | Sets a named control (number, true/false or text) with an optional ramp. |
| Q-SYS snapshot | Loads a snapshot bank and number with an optional ramp. |
Live > Yamaha console actions shows everything Audio X can tell the console to do.
If the console does not give scene names, the recall actions are still listed, without names, and the window says so.
The command paths follow the published Yamaha RCP conventions and were tested against a simulated console. Before a show, use the RCP box in Show control (or this list's Send button) to confirm the scene recall and the master fader path on your own TF, and use a Yamaha raw action for anything that differs. MIDI scene recall depends on the console's Program Change settings.
Control Audio X from a phone, tablet or another PC on the same network.
Network > Remote control: tick Allow control from a phone or tablet.
http://192.168.0.164:8765, and the secure address on the next port (https://...:8766) that you need to use a phone's microphone.Anyone with the PIN can control the show: use it on the show network only. Five wrong PIN attempts lock that address out for a minute. The plain page is not encrypted.
Open the address in a browser, enter the PIN. "Add to Home screen" gives a full-screen app icon. The page keeps the screen awake and reconnects by itself.
GO, STOP, PAUSE and PANIC, the now-playing display with elapsed, remaining and next, the countdown timer, and the cue list: tap a cue to put it on standby, tap its play button to fire it.
Start and stop the radio, see what is playing, HOLD TO TALK (or tick Latch to tap to talk and tap to stop), the Talkback microphone chooser with its level bar and status line, and a button for every ident.
To use this device's own microphone for announcements, open the secure address (https://...:8766), accept the browser's certificate warning once, sign in, and choose This device's microphone in the Talkback microphone list. The page then streams the microphone to Audio X while TALK is on.
All 16 pads of the chosen bank, banks A to D, Stop all pads, and a note of how far the playlist and radio are being ducked. Hold pads play only while pressed.
A view-only copy of the analyser. It shows a message if the analyser is not running (start it in Live > Monitor).
http://<pc>:8765/confidence (or /stage) is a view-only page for a TV or laptop on stage. Use its Tiles button to choose what it shows. See Stage Display.
AudioXRemote-1.0.1.apk is a small Android app that shows these same pages full screen.
| Setup | Sign in | Playlist |
|---|---|---|
 |  |  |
| Radio | Soundboard |
|---|---|
 |  |
If the app cannot connect, it shows which of the two ports answers. Check that Network > Remote control is switched on in Audio X, that both devices are on the same Wi-Fi, and the firewall button.
The stage display (also called the confidence monitor) shows the show's clock, what is playing and what is next, so performers and crew can see it on any screen.
It shows the clock, the cue name, the time remaining with a red warning flash at the end, elapsed time, the next cue, the countdown timer while it runs, the sound level and stereo meters, and your logo and title (see Branding in Settings Reference).
Choose a screen in the Stage box at the top of the main window and press Open. F11 or the window's full-screen button fills the screen.
Network > Remote stage display shares the display with other PCs.
On the other PC install AudioXStage-Setup-1.0.0.exe (or unzip AudioXStage-1.0.0-win-x64.zip). It lists Audio X PCs it finds on the network, or you type the address, and it reconnects by itself after a network drop. F11 or double-click for full screen, F2 or right-click for the connection panel, Esc to leave full screen. For a kiosk screen it can be started with AudioXStage.exe --host 192.168.1.20 --fullscreen --screen 2. Use Allow in Windows Firewall if a Viewer cannot connect.
Any browser (phone, tablet, smart TV, laptop) can show it. In the Watch in a web browser section of the Remote stage display window, copy the address http://<this PC>:5890: it sends the browser on to the stage page of the web control, where you sign in with the web PIN (Network > Remote control must be on). The page's Tiles button chooses what it shows. The Android app has an Open stage display button for the same page.
The viewers can only watch, nothing can be controlled from them. The connection is not encrypted: use it on the show network.
Audio > Tone generator makes test signals for tuning a system.
| Signal | Notes |
|---|---|
| Sine, square, triangle, sawtooth | Band-limited. |
| Log sine sweep | Start, end and duration; repeats. |
| White, pink, brown, blue, violet noise | Pink is flat to within 1 dB from 160 Hz to 16 kHz. |
| Pink noise in a 1/3 or 1-octave band | Centred on the chosen frequency. |
| Pink noise bursts | On and off times with soft edges. |
| Click / impulse train | Clicks per second; level is peak. |
1-2, 1,2,5-8 or all. Noise can be independent on each output or identical.Audio > Surround up-mix turns the stereo programme (cues, radio, loopback on outputs 1-2) into a 5.1 or 7.1 mix on the open device.
This is an independent up-mixer, not a Dolby or DTS decoder, so it will not sound identical to those.
Audio X does not contain a full editor; it hands audio to OpenDaw, a free open-source program that you install separately.
If Audio X cannot find OpenDaw, it offers Open download page and Locate OpenDaw.exe. The generated project files could not be tested in a real OpenDaw; if one opens empty, drag the WAV files from the session folder into OpenDaw. Your WAV files are never changed.
Opens the driver and sets rate, buffer and channels. See Getting Started.
When Audio X starts: open the last show, open the audio device, start the radio (after the device is open), open the stage display on the chosen screen, start full screen.
Main screen: show the row of 10 soundboard pads, show the countdown timer, keep Audio X on top of other windows, ask before closing.
Go full screen now (F11) switches full screen on and off.
GPU and parallel-job settings. See Hardware Acceleration.
Click a key box, then press the key you want (with Ctrl, Alt or Shift if you like). Esc cancels, Delete removes the key. A key can only do one thing. You can map GO (default Space), STOP, PAUSE, PANIC (default Esc), standby next and previous cue, radio start/stop, TALK (hold), the countdown, markers, any of the ten main-row pads, any of the 64 pads, and the 16 pad keys of the soundboard window. Reset all to defaults restores the defaults.
Yamaha, MIDI and Q-SYS equipment. See Show Control.
Sets the application title, the logo (PNG or SVG; Use default restores Audio X's) and the primary and secondary colours. The title, logo and colours appear in the main window and on every stage display. Apply previews, Save & close keeps, Reset restores the defaults.
Shows the version, the copyright (C) 2026 Justin L Worley, the AGPL-3.0 licence notice and the no-warranty statement, the project web site and the contact address (contact@audiox.info). Read the licence opens the full text. See Licence and Copyright.
Tick Set the clock from a network time server to use a corrected time. The PC's own Windows clock is not changed. The corrected time is used by the stage display, its viewers, the radio's clock-time idents and the cue clock. Pick a server from the list (or type any host name), press Test server, then Use this server. For a live show choose a nearby, reliable server such as time.cloudflare.com, time.google.com or a pool.ntp.org region. The time is re-checked every 30 minutes. The NTP tick in the main window's top bar switches it on and off.
Sends Art-Net TimeCode (UDP port 6454) for lighting desks, media servers and show controllers.
SMPTE LTC audio and MIDI timecode are not generated.
Every item of the six menus, in the order they appear. Each item opens a window or does one thing; nothing appears in two menus.
The top bar also has three buttons: + Add Cue(s), Edit Cue and Process List.
| Item | Page |
|---|---|
| Open show | Cues and Shows |
| Save show | Cues and Shows |
| Item | Page |
|---|---|
| Add stems (group mono files into one cue) | Cues and Shows |
| Add action cue (no audio) | Show Control |
| Cue actions (Yamaha / MIDI / Q-SYS) | Show Control |
| Remove cue | Cues and Shows |
| Normalise all cues | Process List and Normalise |
| Item | Page |
|---|---|
| Radio and idents | Radio |
| Soundboard | Soundboard |
| Monitor: spectrum and SPL | Monitor: SPL and RTA |
| Input monitor (Dante / USB levels) | Input Monitor and Recorder |
| Multitrack recorder (up to 48 tracks) | Input Monitor and Recorder |
| Yamaha console actions (scenes, mutes, faders) | Show Control |
| Item | Page |
|---|---|
| Channels and microphones | Microphones and Talkback |
| Surround up-mix (5.1 / 7.1) | Audio Tools |
| Tone generator | Audio Tools |
| Media converter (audio and video, GPU) | Media Converter |
| OpenDaw (DAW / audio editor) | Audio Tools |
| Item | Page |
|---|---|
| Remote control (phone / tablet web page) | Remote Control and the Phone App |
| Remote stage display (other PC) | Stage Display |
| Time server (NTP) | Settings Reference |
| SMPTE timecode (Art-Net) | Settings Reference |
| Item | Page |
|---|---|
| Audio device | Getting Started |
| Preferences (start-up, full screen, layout) | Settings Reference |
| Hardware acceleration (graphics card) | Hardware Acceleration |
| Key mapping (keys for buttons) | Settings Reference |
| Show control (TF5 / Q-SYS / MIDI) | Show Control |
| Branding (logo and title) | Settings Reference |
| About and licence | Licence and Copyright |
| Key | Does |
|---|---|
| Space | GO |
| Esc | PANIC |
| F11 | Full screen on and off |
All other keys can be set in Key mapping.
Dante Virtual Soundcard needs a working wired Dante network before its driver starts. In the Dante Virtual Soundcard window:
The Audio device window shows this checklist and the state of the PC's wired network adapters when the device will not start.
Read the sentence under the level bar in the Radio window or on the phone page.
| It says | Do this |
|---|---|
| The audio device is not open | Open it in Settings > Audio device. |
| Input N is not open on the audio device | Open more Input channels (ASIO only), or choose a PC input device instead. |
| Waiting for audio from the web device | On the phone, open the secure https://...:8766 address, accept the certificate, allow the microphone and choose This device's microphone. |
| No signal / very quiet | Check the microphone is plugged in and not muted; check the right channel. |
http://192.168.0.164:8765. The Android app shows which of the two ports (plain and secure) answers.Install FFmpeg (for example winget install Gyan.FFmpeg) or press Set ffmpeg in the Media Converter and choose ffmpeg.exe. Audio X looks in the path you set, the Audio X folder, the system PATH and the WinGet links folder.
Open Settings > Hardware acceleration. If it lists no working GPU encoder, update the graphics driver and press Look again. Make sure the card is not set to Off. Only H.264, H.265 and AV1 video are encoded on the GPU; audio work runs on the CPU.
Spotify is captured with Windows process loopback. If the levels look uneven, switch off audio enhancements and spatial sound on the playback device.
In %AppData%\AudioX. Shows are the .axshow files you saved. Deleting settings.json returns Audio X to its defaults.
Audio X is copyright (C) 2026 Justin L Worley.
Audio X is free software. It is released under the GNU Affero General Public License, version 3 (AGPL-3.0). This manual is covered by the same notice.
The AGPL has one rule that the ordinary GPL does not. If you change Audio X and let other people use your changed version over a network, you must offer those people the source code of your version. For Audio X this covers the phone and tablet web remote, the stage display and the Android app. If you only run the unchanged program, there is nothing to do.
Audio X comes with no warranty, to the extent the law allows. The author is not liable for damage or loss of any kind arising from its use. In a live venue, rehearse every show and keep a backup way to play audio.
LICENSE.txt (the full licence) and NOTICE.txt (copyright and third-party components) are installed next to AudioX.exe.The project web site, audiox.info, has the downloads and the complete source code as a .zip file (AudioX-1.4.0-source.zip). BUILDING.md inside it explains how to build Audio X.
Questions, bug reports and ideas: contact@audiox.info.
Audio X is built with JUCE, the Steinberg ASIO SDK, .NET and a few NuGet packages. Each stays under its own licence; the list is in NOTICE.txt. FFmpeg is not part of Audio X: you install it yourself and it has its own licence. Yamaha, Dante, Q-SYS, Spotify, NVIDIA, Intel and AMD are trademarks of their owners and are named only to describe what Audio X works with.