UAE-Based CNTXT AI Launches On-Device Arabic Speech-to-Text System ‘Munsit Edge’
Built on the company’s Munsit Arabic automatic speech recognition model, Munsit Edge enables real-time transcription across major Arabic dialects on consumer hardware.
You're reading Entrepreneur Middle East, an international franchise of Entrepreneur Media.
CNTXT AI, a UAE-based data and AI company, has launched Munsit Edge, an Arabic speech-to-text system that runs fully on-device without a cloud connection.
Built on the company’s Munsit Arabic automatic speech recognition model, Munsit Edge enables real-time transcription across major Arabic dialects on consumer hardware.
The system processes audio locally on smartphones, computers, in-car systems and Internet of Things (IoT) devices [devices that can connect to the internet], eliminating the need to send data to external servers.
“Until today, the Arab world has never had Arabic speech recognition that truly ran on the devices people use,” Mohammad Abu Sheikh, CEO of CNTXT AI, said. “With Munsit, we proved you can build an Arabic‑first model that beats generic systems on accuracy. With Munsit Edge, we’ve moved that model out of distant data centers and onto the devices themselves.”
According to CNTXT AI, the system delivers approximately 24% word error rate across major Arabic dialects and supports real-time transcription with about 150-millisecond latency on standard mobile devices.
Munsit Edge is available through software development kits (SDKs) for iOS, Android, macOS, Windows and Linux, on-premise container deployments, and embedded IoT integrations.
CNTXT AI, a UAE-based data and AI company, has launched Munsit Edge, an Arabic speech-to-text system that runs fully on-device without a cloud connection.
Built on the company’s Munsit Arabic automatic speech recognition model, Munsit Edge enables real-time transcription across major Arabic dialects on consumer hardware.
The system processes audio locally on smartphones, computers, in-car systems and Internet of Things (IoT) devices [devices that can connect to the internet], eliminating the need to send data to external servers.
“Until today, the Arab world has never had Arabic speech recognition that truly ran on the devices people use,” Mohammad Abu Sheikh, CEO of CNTXT AI, said. “With Munsit, we proved you can build an Arabic‑first model that beats generic systems on accuracy. With Munsit Edge, we’ve moved that model out of distant data centers and onto the devices themselves.”
According to CNTXT AI, the system delivers approximately 24% word error rate across major Arabic dialects and supports real-time transcription with about 150-millisecond latency on standard mobile devices.
Munsit Edge is available through software development kits (SDKs) for iOS, Android, macOS, Windows and Linux, on-premise container deployments, and embedded IoT integrations.
