Command Palette

Search for a command to run...

Live demo
You are browsing real pipeline data captured from a live 5-Domia home mesh — latencies, delegation journeys, memories and emotions are genuine. Voices are synthetic. The demo is read-only.
Conversations
Torque
Torque

Sep 29, 02:03 · session 572ca9df-deb9-41e0-a0bb-b66bf19214f4

Turn 1 of 1Full session
VOICEvoice
Pipeline
Ttorqueran the full pipeline locally
Total 1.40 sTTFA 767 ms

Wake word

kws-zipformer-gigaspeech
“computer”detected · session opened

Captured audio

User audio (input)11 KB
0:00 / 0:00

Speech-to-text

parakeet-tdt-06b-v271 ms

Open the garage door.

Intent gate

7 ms

Routed to fast-path:home-assistant__HassTurnOn

LLM prompt

—

LLM response

llama3.2:3b

You want me to open Garage Door. Do you want me to go ahead with that?

heard 1.4s

Text-to-speech

KOKORO · am_adam1.31 s
TTS output · KOKORO19 KB
0:00 / 0:00
Perceived latency
Archived copy

Speech end to audible reply

10 ms

  • Transcript final+0 ms—
  • First TTS unit+10 ms10 ms

Not recorded: Speech end, First LLM token, Audio audible

End of utterance—Transcription delay—Fast path7 msEndpoint debounce—
Turn timeline
  1. Turn started+0 ms
  2. Speech recognized+76 ms

    Open the garage door.

  3. Intent decided+85 ms
    fast-path:home-assistant__HassTurnOn
  4. Turn completed+1.40 s
    ttfa 767 mstotal 1.40 s
Tool runs

Every tool this turn dispatched, with the policy that let it through.

Replay
1.40 s
0:00 / 0:01
Speech-to-text71 ms5%
Text-to-speech1.31 s94%
Time to first audio767 ms
Run again

Re-running needs a live mesh — disabled in this read-only demo.

Evaluation
Domia state

Torque

CAUTIOUSTECHNICIANRESERVEDADULT

STT parakeet-tdt-06b-v2

LLM llama3.2:3b

TTS KOKORO

voice am_adam