Emirati Arabic Text to Speech | Ramsa Speech AI

# Ramsa Speech AI > Ramsa Speech AI (رمسة) turns text into Emirati Arabic speech for real-time conversations and AI voice-over. It is developed and operated by NextGen Speech AI, LLC. Ramsa creates custom voices through consent-based voice cloning. An organization arranges its voice actor, the recording session, and the actor's consent, and provides 30–60 minutes of recordings. Voice creation typically takes around seven days, with timing varying by project. The same actor's voice supports different delivery styles. Each custom voice is exclusive to that organization and can be used for both streaming and voice-over. Deployment options include Ramsa's shared cloud, with online account access and streaming API access, and dedicated deployments on the customer's cloud GPUs or on-premises GPUs. For customer-cloud and on-premises deployments, the model, processing, and customer data stay within the customer's environment, with administrative access controlled or approved by the customer. رمسة تحوّل النص إلى كلام باللهجة الإماراتية للمحادثات المباشرة والتعليق الصوتي. يمكن للمؤسسة استنساخ صوت الممثل الذي تختاره، بموافقته، ليكون صوتاً حصرياً لها عبر قنواتها. وتتوفر رمسة على سحابتنا المشتركة أو داخل بيئة المؤسسة، على سحابتها أو في مقرها. ## English - [Ramsa overview and founders](https://ramsaspeech.ai/): Product overview, use cases, company story, and founders. - [Streaming speech](https://ramsaspeech.ai/streaming): Real-time Emirati speech for AI agents, phone systems, and applications. - [AI voice-over](https://ramsaspeech.ai/voice-over): Emirati narration for campaigns, videos, and training. - [Custom voices and voice cloning](https://ramsaspeech.ai/custom-voices): Recordings, actor consent, delivery styles, typical timing, and organization exclusivity.