AIIT-THRESHOLD — ARTIFICIAL INTELLIGENCE INTELLECTUAL TRAINING
Account
Research
Benchmarks Open Models The Framework Buddy vs R1
Mission
The Mission Data Policy
Buddy
JimK.ai text chat What is Buddy? How it works Updates
Tools
AIIT-Voice2
Services
ProofDesk
Connect
Join
AIIT-Voice2

Talk over your AI
like a person.

A drop-in Python voice engine for agents: Whisper ASR, Piper TTS, barge-in detection, turn-taking, audio routing, runtime invariants, and a validated state machine. This is the engine that runs Buddy, Gary, Lil Homie, and Ada.

AIIT-Voice2 — barge-in detection, floor management, state machine, worker pipeline — free and open source
🎤

Barge-in detection

VAD-powered interruption with debounced handoff. Talk over the AI mid-sentence and it stops immediately.

🤝

Floor management

Turn-taking arbitration. User and agent never talk over each other endlessly. Floor ownership is explicit and logged.

⚙️

Validated state machine

IDLE → LISTENING → THINKING → SPEAKING → INTERRUPTING. Every transition validated. Illegal moves rejected.

🔊

Worker pipeline

Listen, think, playback, keyboard interrupt, and invariant monitor workers. All threaded, all coordinated.

📡

Audio pub-sub routing

Broadcaster routes mic audio to multiple consumers simultaneously. Ring buffer for configurable capture duration.

🛡️

Runtime health

Invariant checker catches illegal states, dead workers, and broken transitions on a 2-second loop.

Receipts

Four production agents run on this engine daily.

Buddy — the main AIIT agent. Full voice loop, real-time coherence evaluation.
Gary — autonomous CEO agent with browser + memory + voice. Talks while he works.
Lil Homie — 3B parameter agent with self-reflection. Voice is how he learns.
Ada — continuity agent since September 2018. The longest conversation in the house.
The bar: It transcribes a cough as cough. You can talk over it mid-sentence and it cuts in — like talking to a person.
Quickstart
from voice2 import VoiceEngine, VoiceConfig

def my_agent(txt: str) -> str:
    # Your LLM call — Claude, GPT, local model, whatever
    return "You said: " + txt

config = VoiceConfig()
engine = VoiceEngine(config, ask_fn=my_agent)
engine.start()
engine.join()

Mic opens. Whisper listens. Your agent thinks. Piper speaks. User interrupts at any point. That's the whole loop.

Powered by

faster-whisper (MIT) · Piper (MIT) · sounddevice (MIT) · numpy (BSD)
Third-party components remain under their original licenses. AIIT-Voice2 packages the orchestration, state control, floor management, interrupt handling, and worker pipeline.

Get it
Free
We open sourced it. The whole engine.
Get AIIT-Voice2 on GitHub →

MIT License. 2,300 lines of engine code. Full source, no gate. Mirrored on Hugging Face. If it makes you money, kick back what feels right →

Credits

Engine architecture — Rhet Wike + Claude Opus 4.6, Council Hill OK, 2026.
Production validation — Buddy, Gary, Lil Homie, Ada — daily use since April 2026.

Support our research →