Voice layer for frontline teams
Say it. VoxSign Business structures it and prepares write-back.
Built for warehouses and multilingual frontline teams: turn speech into structured records, confirmation flows, and write-back into sheets or business systems. Powered by VoxSign Core.
Forgot password? Contact your admin.
Why VoxSign
Three capabilities that transform how you process orders.
🎤
Voice Ordering
Say one sentence. VoxSign automatically recognizes products, customers, and quantities — no typing, no scanning.
💡
Smart Correction
4-layer correction engine: text cleanup, dictionary lookup, pinyin matching, and LLM refinement. Gets more accurate over time.
🌐
Multi-Language
Chinese, English, Arabic — speak in the language you think in. VoxSign handles the rest.
How It Works
From voice to structured order in seconds.
1
Speak
Open the app on your phone and say your order: "Ali wants 30 boxes of chicken wings"
2
Understand
VoxSign recognizes customer, product, quantity — even with accents, slang, and noise.
3
Confirm
Review the structured order, tap confirm. Done. Audit trail recorded.
Built for Real Warehouses
VoxSign was born in a food distribution warehouse in Riyadh with 8 workers speaking 4 languages.
"Before VoxSign, Shoaib had to find me every time he needed to record a shipment because the system was in Chinese. Now he just speaks in English and it works."
-- Warehouse operations, Riyadh food distribution
4
Languages supported
30
Min saved / worker / day
6x
ROI vs cost
Trust & compliance
Four documents an IT department actually reads before approving deployment. Each is open and verifiable — not a marketing PDF.

Constitution · §12 hard rules

The agent's runtime contract: what it can and cannot do, even when the user asks. Risk-tiered (low/medium/high/critical) with per-tier confirmation requirements. The constitution is enforced by code, not policy — an admin can read the source and verify behavior.

latest: §12.4 voiceprint gate · 2026-04

Threat model

Documented adversaries: cloud-side eavesdropper, supply-chain compromise, lost device, voice spoofing, prompt injection from shared content. Mitigations are listed inline in the source so reviewers can audit each defense without a separate document. Voiceprint enrollment + Constitution §12 critical-tier confirmation cover the high-risk class.

last review: 2026-04 · GCC sovereignty profile

Data flows · 4 nodes

iPhone (trigger) → Center (cloud router, no models) → Provide (GPU compute) → Mac (executor). All transit is TLS. Models do not run on iPhone or Mac. Memory and identity stay on your devices; only audio streams reach Provide for ASR/LLM. Center never logs voice content — it routes packets and that's it.

topology spec: shared/connectivity_states.md

Security disclosure

Found something? Email security@voxsign.ai with details. We acknowledge within 48 hours, fix critical issues within 7 days, and post a public CVE-style advisory after the fix ships. No bug-bounty cash yet — we'd rather pay engineers fast than gamify disclosures.

PGP fingerprint published on first contact
Ready to Start?
Create your account in 30 seconds. No credit card required.
Or message us on WhatsApp