ConvoZen.AI, the enterprise conversational AI agentic platform of NoBroker, has expanded into West Asia with the launch of Alif and Rawi, two proprietary speech models purpose-built for the Arabic language as it is actually spoken: bilingual, multi-dialect, and heavily code-switched with English.
Alif is ConvoZen.AI’s bilingual Arabic-English speech-to-text model, engineered to transcribe the messy, multi-dialect, code-switched reality of everyday Middle East and North Africa (MENA) conversations rather than the clean, formal Arabic on which most commercial automatic speech recognition (ASR) systems are trained.
Rawi is a native Arabic code-switching text-to-speech model, addressing the speaking side of the same problem. It is engineered to speak with the natural rhythm, intonation and multi-dialect flexibility that real-world business-to-consumer (B2C) and enterprise voice interactions demand.
“Every market we’ve taken our speech models into has taught us the same lesson: People don’t speak in the clean, formal register that most voice AI is trained on. They mix languages mid-sentence, they carry regional accents, and they expect a system that keeps up. We saw that in India with Akshara and Ragini, and we’re seeing the exact same pattern across the Middle East. Alif and Rawi aren’t an adaptation of an existing model for a new market. They’re built from the ground up for the way Arabic is actually spoken,” said Akhil Gupta, co-founder and chief technology officer of NoBroker.
The launch comes months after ConvoZen launched its conversational AI stack, along with frontier speech models, in India. The two models, Akshara (speech-to-text) and Ragini (text-to-speech), were built to address the conversational infrastructure challenges faced by NoBroker. ConvoZen integrates the full lifecycle into a single platform built on conversational AI agents, copilot AI agents, supervisor AI agents and customer AI agents.
The two models are now deployed across more than 40 enterprises in nine Indian languages, including HDFC Bank, Tata AIG, Cars24, Apollo and Maruti Suzuki.
Ragini, which works across six languages, is tailored for banking, financial services and insurance (BFSI), automotive, healthcare, direct-to-consumer (D2C) and edtech deployments.
