Audio Input

Triggerstriggercore.trigger.audio_in

Send a microphone clip into the live graph while you hold to talk. Emits the audio and a trigger, the way Chat Input emits text. Feed the audio into an STT node to transcribe it.

Trigger. It starts turns on its own: when it fires, the nodes wired to its outputs run.

Outputs

Port	Type	Notes
trigger	event
audio	audio
lang	lang	Optional

Knobs

Right-click a knob and pick Convert to input to drive it from a wire.

Knob	Kind	Default	Notes
placeholder	text	Hold to talk...

Example

The editor has no microphone button for this node yet: a clip comes from another program while the graph is On. It posts JSON to /audio/<graph>/<node id> with the install token:

curl -X POST http://127.0.0.1:8770/audio/voice/Mic \
  -H "Authorization: Bearer $(cat user/data/token)" \
  -H "Content-Type: application/json" \
  -d '{"audio": "data:audio/wav;base64,UklGR...", "lang": ""}'

Wire audio into an STT node and trigger into its trigger. The Voice guide has the whole graph.

More in Triggers

• Agenda: Fire from data instead of a fixed clock: poll a wired store table of {when, payload} rows and fire each row when its when is due.
• Chat Input: Type a message and send it into the live graph.
• Interval: Fire an event every N seconds.
• Manual: Fires once when the graph turns On, and again whenever you press its Fire button.
• Schedule: Fire on a cron schedule: ‘min hour day month weekday’, in the zone its timezone knob picks.
• Webhook: Listen for HTTP calls from outside.