- New system combines speech recognition and voice generation, allowing organizations to process and use Arabic voice at scale
Abu Dhabi, UAE. April 2026: As the UAE accelerates its investment in artificial intelligence and digital services, Abu Dhabi based CNTXT AI has expanded Munsit into a unified Arabic voice AI platform, combining speech recognition and voice generation in a single system designed for real-world use across government, finance, and digital platforms.
The update comes at a time when voice is becoming an increasingly important interface for artificial intelligence globally, while demand for Arabic language solutions continues to grow across the region. From customer support and digital banking to government services and media, organizations are handling rising volumes of spoken interactions but have had limited tools to process them accurately.
CNTXT AI says Munsit was built to address that gap.
“We are done being spectators in the AI race. Arabic has been an afterthought in global AI for too long,” said Mohammad Abu Sheikh, Founder and CEO of CNTXT AI. “Munsit is proof that when you build for the language from day one, you don’t just compete, you lead. This is the world’s most accurate Arabic voice AI, built here, for our region, and designed to power what comes next.”
Originally launched as a speech-to-text model, Munsit has now evolved into a full voice engine. The latest version combines CNTXT’s patented speech recognition technology with its new Faseeh text-to-speech model, allowing systems not only to understand Arabic speech but also generate it naturally across more than 25 dialects.
This creates a closed loop system where audio can be captured, transcribed, analysed, and converted back into natural spoken Arabic within the same workflow.
The platform is available through API, web workspace, and mobile app, and can be deployed on cloud, private infrastructure, or fully on premise, allowing organizations to maintain control over sensitive data.
Early adoption has been strong. CNTXT AI says Munsit has already processed more than 86 million Arabic words and over one million minutes of audio, with more than 250 government and enterprise organizations using the system. Its mobile application has also reached over 150,000 users within two months.
The company says this level of usage reflects a broader shift in how AI is being applied across the region.
As ecommerce, digital banking, and online services expand in the UAE and wider Middle East, the volume of voice interactions has increased significantly. Until recently, much of that data remained difficult to analyse due to limitations in Arabic speech recognition, particularly across dialects.
Munsit allows organizations to work with that data more effectively.
A call center can analyse all customer interactions instead of reviewing small samples. Financial institutions can use voice data for compliance and quality control at scale. Government entities can transcribe and search meetings and citizen interactions in the dialect they were spoken, while keeping data within their own infrastructure.
For developers, the platform provides a way to integrate Arabic voice capabilities directly into applications without building models from scratch, reducing both cost and time to market.
The launch also reflects the UAE’s broader focus on building local AI capabilities. While global technology companies continue to dominate the underlying model layer, regional players are increasingly developing systems tailored to local languages and operational needs.
CNTXT AI says its goal is to position Munsit as part of that ecosystem.
“Arabic has been underserved in AI for a long time,” Abu Sheikh said. “If the region wants to build serious technology, it needs systems that understand its language properly. That’s what we’re building.”
The new version of Munsit is now available via app.munsit.com.