Pulse 112

AI emergency dispatch with real-time voice triage and emotion analysis at sub-second latency — Hume AI flags caller distress, Mappls routes the nearest unit. Built to cut initial triage time by 40%.

TYPE
AI emergency dispatch
ROLE
Lead developer
BUILT
Nov 2025
STATUS
Hackathon build · Live demo
VISIT
112-emergency.vercel.app
SOURCE
GitHub
TECH STACK
Next.jsTypeScriptOpenAITailwind CSSHume AIMappls
Pulse 112 screenshot
01

The first minute matters most

In an emergency call, dispatchers have to judge urgency from a voice that's often panicked. Pulse112 listens alongside them and turns the call into a structured, prioritised incident in real time.

02

Hearing distress, not just words

Hume AI analyses vocal emotion so the system can flag fear and stress that the transcript alone would miss. OpenAI handles triage questions, and both run with sub-second latency.

  • Emotion-aware priority scoring
  • Live transcript and suggested questions
  • Incident cards dispatchers can act on immediately
03

Sending the nearest help

Mappls provides location and routing, so the dashboard can show the caller's position and route the closest available unit.

04

Results

Designed to cut initial triage time by 40%Sub-second voice analysisLive product demo and pitch deck
FAQ

Questions people ask

Does it replace dispatchers?

No. It's an assistant: it surfaces urgency and location faster so the human makes the call sooner.

NEXT CASE STUDYGramUdyam SaathiVoice-first loan advisor for first-time rural entrepreneurs under MoSJE schemesRead it

GOT AN IDEA WORTH BUILDING?
LET’S TALK IT THROUGH

Start a conversation

Open to full-time roles, internships and freelance builds.

Tell me what you’re working on.