01
A roaming mic breaks the moment.
For an audience member to speak, someone crosses the room with a microphone. The pace drops, and so does the magic.
Perceptive sound systemSound that
Sound that
sees.
Briquemault links a venue’s cameras, microphones, speakers and lights into a single system. It sees what happens on stage and in the audience, and puts every sound and every light in the right place, at the right moment.
In developmentSimulation
Designed to integrate with
- L-ISA
- ADM-OSC
- PosiStageNet
- OSC
- Lighting · DMX, Art-Net, sACN
- PTZ cameras · VISCA, NDI
- Beamforming ceiling mics
The problem
Venues have the hardware. What they lack is the brain that connects it.
02
Sound ignores the stage.
Sounds follow a timeline, not what actually happens. When the performer improvises or moves, the sound stays where it was.
03
Immersive sound takes a team of experts.
Placing dozens of sound objects by hand, cue by cue, takes days of work and a specialist at every performance.
The solution
Five functions, one system.
Briquemault does not replace your hardware. It makes it work together, at the pace of the show.
- 01See
Cameras that understand the stage.
Fixed or PTZ cameras, in infrared when it is dark, track performers with nothing to wear, locate the audience as anonymous positions and detect key moments: an object raised, a gesture, an entrance.
- Markerless performer tracking
- Anonymous audience positions, no facial recognition
- Invisible infrared for dark rooms
CAM 01 · VISIBLESimulation - 02Hear
The audience speaks, no microphone needed.
Briquemault locates in 3D the volunteer the camera designates and sends that point to a steerable ceiling mic array, which aims its beam at them. They speak from their seat, and the whole room hears them.
- The volunteer’s 3D position, computed and sent to the mic
- Voice isolated and cleaned locally, feedback under control
- No more microphone to pass around
Simulation - 03Place
Every sound in its real place.
Briquemault drives your immersive sound system over ADM-OSC. An audience member’s voice comes from their seat, an effect crosses the room, the sound follows the performer into the audience.
- Sound objects driven over ADM-OSC and OSC
- Performer positions over PosiStageNet
- Ultra-directional speakers for a sound only one person hears
Simulation - 04Understand
An AI that knows the cue list.
Claude, Anthropic’s AI, reads the show’s cue list and connects what the cameras see to the right sound cues. When the performer improvises, it adapts. The sound operator talks to it like to a stage manager.
- Decisions at cue level, on the scale of seconds
- Handles the unexpected and improvisation
- Conversational control: “next act, the voice comes from the ceiling”
Simulation - 05Direct
A control room that follows the action.
Lights follow the performers and PTZ cameras frame the action on their own. The same installation films the show from several angles.
- Lights driven over DMX, Art-Net or sACN, following the action
- Automatic framing for PTZ cameras
- Multi-angle capture of the show
- Manual override at any time
REC · AUTO FRAMINGSimulation
In a magic show
Effects you cannot do with a microphone and two speakers.
The whisper
A voice whispers a clue that only one audience member hears. The whole room watches them react.
The travelling object
The card vanishes from the magician’s hand. Its sound flies over the audience and lands in row 8.
The room has ears
Anyone can answer. Their voice fills the room, and it comes from them.
The reveal from the seat
The mentalist guesses a word, and the sound of the reveal comes from the exact seat of the spectator.
The sound that follows
The performer walks into the audience. Their voice follows, row after row.
Imagine anything
The possibilities are endless.
A room that sees, hears and understands its show becomes an instrument. Here are a few ideas; yours will go further.
- 01
The moving rain
The storm breaks above the front row, then spreads across the whole room, row by row.
- 02
The ghost in the rows
Invisible footsteps cross the audience, stop behind one person, then move on.
- 03
The room that votes
The audience raises hands: the camera counts, and the show picks what comes next, live.
- 04
A duet with the ceiling
A voice answers the performer from the ceiling, picked live from recorded lines according to what they do.
- 05
The orchestra that follows
In concert, each instrument sounds from where the musician stands, even as they move.
- 06
The shared secret
Two audience members, at opposite ends of the room, hear the same sentence. Nobody else does.
- 07
The living conference
A participant’s question starts from their seat, and the speaker’s answer travels back to them.
- 08
The museum that listens
Each visitor triggers the story of the work they stop in front of, with no headset or button.
Nothing is fixed: each show builds its own effects in its cue list, combining what the cameras see, what the mics hear and where every sound sits. The operator tunes them in plain language.
Why Briquemault
Designed for interaction, not just for playback.
The camera locates, the mic aims
Briquemault locates the volunteer in 3D and sends that point to a steerable ceiling mic array, which aims its beam at them. Their voice is then rendered from their seat in the immersive system.
The AI picks the cue, live
Claude connects what the cameras see to the cue list and arms the right cue, at one of three autonomy levels: suggestion, operator approval, or automatic on authorised cues.
Nothing to wear
No tags on performers: the venue’s cameras do the tracking. Trackers already installed still work over PosiStageNet.
Built for blackouts
Infrared invisible to the audience: tracking carries on when the lights go out, exactly when the magic happens.
Hardware-agnostic
ADM-OSC, OSC and PosiStageNet in and out, to drive the hardware a venue already owns.
Anonymous and safe by default
Positions computed on site, no images kept, no facial recognition and no emotion reading. If the AI drops out, the regular cue list carries on.
The test venue
A full-scale test venue, with a real audience.
Briquemault is developed at the Domaine de Briquemault, in the Loiret, in the Crystal Orangery: 400 m² under glass, already rigged with sound and lighting trusses, hosting conferences, seminars, dinners and evening events.
- under glass
- 400 m²
- seated dinner guests
- 250
- standing
- 300
- from Paris
- 1h30
Technical questions
Your objections, our answers.
Briquemault is in development: these answers describe its design principles.
What about latency?
Two clocks, never mixed up. Claude decides at cue level, in seconds, before the sound is needed. Tracking and sound rendering run in a local engine, in milliseconds. The AI is never in the audio loop.
The room is dark.
The cameras work in infrared, with infrared illuminators the audience cannot see. Show lighting does not disturb tracking.
What about feedback when an audience member speaks?
It is the hardest part, and it is treated as a starting constraint: a narrow pickup beam, speakers near the speaking audience member turned down while they talk, a gain ceiling measured at installation, and a capture-only mode when the room does not allow reinforcement.
What if something fails mid-show?
The show goes on. Claude acts at the autonomy level the operator chooses: suggestion, approval, or automatic on authorised cues. Every cue has a default behaviour, the operator can take over at any time, and if the AI or the network drops, the local engine carries on with the planned cue list.
Is filming the audience allowed?
The system is designed for privacy: on-site processing, anonymous positions, no facial recognition and no emotion reading, no audience recording by default, a mic that only opens for the chosen moment, and clear notice at the entrance. Every installation goes through a compliance review.
Could the cameras give our secrets away?
No. Images are analysed on site, then forgotten: nothing is recorded or sent. Only positions and timestamped events come out. Nothing that could reveal a method ever leaves the room.
Do we need to replace our sound system?
No. Briquemault is designed to drive object-based immersive systems over ADM-OSC, including L-ISA, and to adapt to what is already there. A venue can start simple and grow.
Does it need an internet connection?
Tracking and sound run locally. The connection is used for preparation and operator assistance: losing it does not stop the show.
How much does it cost?
A licence per venue, and a quote for installation when the venue is complex. The simplest is to tell us about your space.
Offers
Software. And us, when you need us.
Briquemault licence
The software, per venue, with updates and support.
- All five functions
- Continuous updates
- Operator support
Custom installation
For complex venues, we come to install, calibrate and train your team. We can also bring the hardware for an on-site trial, with you.
- Venue and hardware survey
- Installation and calibration
- Training and opening-night support
Demonstration
At Domaine de Briquemault, in the Loiret, on an L-ISA system. By appointment, as development progresses.
- 1h30 from Paris
- By appointment
“The more the world fills with generated images, the more precious reality becomes. Briquemault does not replace the show with AI: it puts AI at the service of what is real, and makes it invisible.”
The team
The people who needed the tool are building it.
Magicians who have run the Agence des Magiciens since 1994, and a technical partner.
MagicianMao la Magicienne
Marie-Odile Langloys, artistic director of the Agence des Magiciens
LinkedIn
Partners and sponsors
They support Briquemault.
- Temps PastelAgence des Magiciens and Les Décalés
- Empreinte Production
Story
Magicians have always invented the technology of illusion.
Robert-Houdin put electricity to work in his tricks. Méliès, an illusionist, invented the special effects of cinema. Briquemault comes from the same family: magicians building the tool they were missing.
Read the story1845
Robert-Houdin
1888
Méliès
Today
Briquemault