Startup Fundraising

Maya Research Raises $1.9M Seed Funding for Voice AI

Maya Research secures $1.9M seed round from South Park Commons to develop native conversational AI, targeting billions worldwide via voice interfaces.

Share:
AM
Alvaro de la Maza

Partner at Aninver

Stay ahead of the market

Get instant notifications when new news matching "Artificial Intelligence (AI), Technology, Software & Gaming in India" are published.

Key Takeaways

  • Maya Research raised $1.9M (Seed) from South Park Commons.
  • Sector: Artificial Intelligence (AI), Technology, Software & Gaming.
  • Geography: India.

Analysis

Maya Research, an ambitious AI firm focused on developing conversational interfaces for a global audience, has successfully closed a $1.9 million seed funding round. The investment was spearheaded by prominent venture capital firm South Park Commons, signaling strong confidence in Maya's vision to bridge the digital divide through intuitive voice technology.

The company is dedicated to creating AI that mirrors human conversation, moving beyond simple transcription and translation to understand and adapt to diverse linguistic and cultural nuances. This approach is particularly critical for markets where voice is the primary mode of interaction with technology, a segment often underserved by current AI solutions. Maya Research aims to empower billions by enabling them to engage with digital services through natural, spoken language, rather than relying on complex text-based interfaces.

“India is fundamentally a conversational economy, and the current AI discourse has largely overlooked the critical need for localized voice models,” stated Dheemanth Reddy, Co-founder and CEO of Maya Research. “We are building AI that speaks, thinks, and responds natively, ensuring that the next wave of global internet users can access powerful technology in their own language and context. Our goal is for this foundational AI experience to originate from India and serve the world.”

Maya Research has already demonstrated significant traction since its inception in 2025. The company's open-source voice model, Maya 1, has achieved remarkable recognition, ranking sixth globally among open-weight models on the Speech Arena platform, a feat that places it on par with industry leaders like OpenAI's GPT-Realtime-2. This technical achievement is complemented by substantial user adoption, with over 440,000 model downloads on Hugging Face and more than 3 million app downloads across India, Southeast Asia, and the MENA region. This widespread usage creates a valuable data flywheel, accelerating the development of more sophisticated and contextually aware conversational AI.

The strategic importance of voice-first AI cannot be overstated. As billions of new users come online, particularly in emerging economies, the ability to interact naturally via speech will be paramount. Maya Research's focus on building deeply localized and emotionally intelligent voice models positions it to capture a significant share of this rapidly expanding market. The company’s founders, Dheemanth Reddy and Bharath Kumar Kakumani, emphasize that technology should feel inherently accessible and personal, reflecting the user's own communication style.

Prateek Mehta, General Partner at South Park Commons, highlighted the transformative potential of Maya's work. “The internet’s foundation in English and text has inadvertently excluded vast populations. Maya is fundamentally reshaping this by centering communication around speech and local languages. The entity that masters multilingual speech data and robust conversational models will hold a decisive advantage as billions adopt voice interfaces. Maya's early success, evidenced by millions of app downloads and extensive model usage, underscores this potential.” Aditya Agarwal, also a General Partner at South Park Commons, added, “Maya represents India’s pivotal moment in the AI revolution, reimagining access for its population through intuitive modalities.”