Coming soon

About This Game

Talk well and the room leans in. Go flat and they drift. Lose them and the phones come out, one by one, like a slow-motion mutiny. Nobody here is scripted to clap politely. They're listening. Really.

You're a speaker for hire with no reputation and a slide deck that says HOW TO PRESENT GOOD!! Work your way up:

• Monday meeting — rows of colleagues, and the latecomer who always takes the chair in the middle.
  
• Boardroom — ten people, one long table, and a boss who cuts in with "So roughly what's the budget?" mid-sentence.
  
• Wedding banquet — the best man lost his voice. Half the room has their back to you. The other half has wine.
  
• 60-second elevator pitch — one investor, thirty-eight floors. Go quiet for two seconds and he starts checking his watch out loud.
  
• The big stage — sixty seats in the dark, a red circle under your feet. Good luck.

Every gig has a brief, three objectives and a fee. Stars decide how much of it you keep. Harder difficulty pays more. Turning on the helpers costs you a cut.

He goes off his chair. The room goes quiet. Then he files a complaint, your fee is halved and your reputation takes the hit. Worth it? Your call.

When you finish, hands go up — and the questions are about what you actually said. Answer into the mic. They react to that too. It all runs on your own machine.

Sixteen speaker gestures, ten hotkeys, one shop. Nothing says "thought leader" like a purchased open-palm.

Every talk is recorded as a replay: your avatar, your voice, the room's mood, from three camera angles, with a timeline that marks exactly where you lost them.

Attention over time, volume, pace, dead air, where your eyes went, a full transcript, and an exact count of every "um", "like" and "you know". It's a game, but the score is real. No power-up can fake it.

Free practice has every venue unlocked from the start. Bring your own slides (PDF on Mac, PNG/JPG everywhere), or draw an impromptu topic — "Sell us this pen" — and go. Write your own gig with your own talking points and the room will check whether you hit them.

Speech recognition, the audience's questions, your recordings and your reports all stay on your computer. The optional AI audience model is a one-time 1.1 GB download; the game is fully playable without it.

Understands English, Cantonese and Mandarin. Interface in English and Traditional Chinese.

Made by Playwithspect. Playful at heart, respectful in craft.

AI Generated Content Disclosure

The developers describe how their game uses AI Generated Content like this:

After a talk, audience questions are generated at runtime by a small open-weight language model (Qwen3) running locally on the player's computer, based on a transcript of what the player just said. Speech-to-text also runs locally (whisper.cpp). Nothing is sent to any server.