Your friendlyconversational AIcompanion

صوتٌ يفهمُ العربية — كما تُقال.

TTS, STT, dubbing, and live voice in one studio. Ship with Python or Node on the same API.

Free tier · no card required
Open source10,000 chars free / monthNo credit card requiredAPI & SDK ready
20+
voices
8
dialects
Live
agents
🇸🇦MSA
🇪🇬Egyptian
🇱🇧Levantine
🇦🇪Gulf
🇲🇦Maghrebi
🇵🇸Palestinian
🇮🇶Iraqi
RTF 0.09
Nour
MSA
Free tier · StudioLive demo
FastAPI · Modal GPUs · Python · Node
Trusted by teams atFrom startups to sovereigns — 1,200+ teams build with Nur
Aramco
STC
PIF
Careem
Emirates
RBC
SAP
Oracle
Salesforce
Notion
Mada
Ooredoo
Etisalat
Anghami
Talabat
86ms
first-token latency
32
languages
6h+
session length
context window
99.99%
uptime SLA
01 · NUR AGENTS

Agents that actually talk.

Real-time voice agents your customers can interrupt.

Sub-90ms first-token latency, full-duplex transport, and persistent memory across sessions. Drop into telephony, web, or apps in 4 lines of code.

86ms p50
first-token, on-device
32 languages
+ 8 Arabic dialects
Memory
persistent across sessions
Telephony
SIP, Twilio, native
NA
Nur Agents
Active
Customer support
Sales triage
Reservations
Renewals
Voices
Noor — warm
Rami — clear
Mira — bright
Live · Customer support
p50 86ms · 24khz
Hey, I'm trying to change the date on my booking for Friday.
● NOOR · VOICE
Hi Sarah — I can move that. Looking at flight LX317. What date works better for you?
Actually... can you just push it to next Tuesday? Same time.
● NOOR · VOICE
Done. You're now on LX317 Tuesday the 14th, 6:40am. I sent the new boarding pass.
0:14
02 · NUR VOICE

Characters who improvise.

Voices, vision, and memory for game worlds.

Per-character emotion vectors, in-engine plugins for Unity and Unreal, lip-sync, and prosody control. Build NPCs that surprise the player — every time.

Emotion vectors
real-time prosody
Unity & Unreal
one-line drop-in
Lip-sync
frame-accurate
Per-NPC memory
scene or persistent
● NUR.gameenginescene · deep_citadel.scn
NPCs: 12 · Voices live: 4
MIRA · NPC
[improvised]
"Stay close. The light here doesn't reach as far as the wind does."
Persona
NameMira
VoiceMira-bright
LangEN, AR
Memoryscene
Emotion
Trust0.62
Fear0.48
Toneurgent
Latency
ttf84ms
rtf0.21x
03 · NUR STREAM

Conversations that last.

Long-session voice for embodied systems.

6+ hour sessions with no degradation. Agentic tool use, on-device inference, and end-of-shift memory commits. Built for fleets that talk for themselves.

6h+ sessions
no quality drift
On-device
edge inference
Agentic
tool use & summarization
Audit trail
every turn logged
● NUR-R1 · OPS CONSOLEfleet · 14 active · 02:11:46 uptime
00:00:12OPBegin warehouse round. Aisle B.
00:00:18R1Confirmed. Beginning aisle B sweep.
00:14:02R1Pallet 14B mismatch. Opening ticket.
00:42:51OPPause. Visitor in zone 3.
00:42:54R1Holding. Handing off to floor lead.
01:33:09R1Resuming sweep. Shift change in 26 min.
02:11:46R13 anomalies tonight. Want a summary?
Session uptime
2.19hrs
Turns
1,432
Memory persisted
Voice activity
The atlas of voices

Built in Riyadh.
Spoken everywhere.

8 Arabic dialects. 32 languages. Hand-tuned in the cities that speak them — not approximated from a transcript pile.

◐ ATLAS · 8 regions tuned
● 01 · field recordingالرياض

Riyadh

Dialect
Najdi
System voice
Noor
Primary use
Government services
Latency · p50
78ms
waveform · sample
Customer story

"We replaced four call-center vendors with one Nur deployment. Our customers stopped asking for a human — and started asking for Noor by name."

RA
Reem Al-Saud
VP Customer Operations · Careem
Read story →
● Manifestoصوت يفكر

For 70 years, machines listened.
Now they're finally
ready to يتحدثوا.

Voice was the most human interface and the worst computer one — stilted, slow, embarrassed by its own latency. Nur was built by people who'd given up on it, then refused to. We are open-source at the core, fluent in 32 languages, and obsessed with the small space between a question and an answer.

signedThe Nur team
RiyadhBeirutSan FranciscoCairo
Try it now · No signup

Hear it. Then ship it.

Pick a scenario, pick a voice, generate. Powered by the same model that ships in production.

Scenario
Voice
Script
● live output
ttf: 86ms·Noor
Press generate to hear me.
24khz · pcm · streamingidle
Pricing

Honest meters. No platform tax.

Same model and SDK at every tier. Open-source core stays free, forever.

Free
$0forever
10k chars · 10m STT
  • Open SDK
  • Community support
  • 60 req/min
Choose
Dev
$29/month
250k chars · 5h STT
  • Voice cloning
  • Live convo
  • Memory 7d
Choose
★ MOST POPULAR
Pro
$99/month
2M chars · 20h STT
  • Multimodal
  • Game SDK
  • Memory 90d
Go Pro
Business
Custom
Unlimited
  • On-device deploy
  • Robotics
  • Memory ∞
  • SLA 99.99%
Talk to sales
Begin

Voice that speaks for you.

Start Building  →Talk to usصوتك للعالم