Gnani AI Raises $10M Series B for Sovereign Voice Agents
Voice is the most natural interface for human-agent interaction, and India is building the infrastructure to own it.
Gnani.ai, a Bengaluru-based voice-first agentic AI startup, raised $10 million in its ongoing Series B round led by Aavishkaar Capital, with existing investor Info Edge Ventures also participating. The announcement dropped on March 31, 2026.
Here's what makes Gnani different from the crowd of voice AI startups. They built the entire stack in-house — speech recognition, voice synthesis, and a proprietary 14 billion-parameter speech-to-speech model. No dependency on OpenAI Whisper or third-party TTS. This is sovereign AI infrastructure, designed to work across 12+ languages with the kind of latency and accuracy that enterprise clients in banking, healthcare, and insurance demand.
The scale is real. 30 million voice interactions processed daily across 200+ enterprise clients. Founded in 2016 by Gopalan and Ananth Nagaraj, this isn't a startup that raised on a pitch deck — they've been grinding for a decade.
The funding will go toward global expansion, advancing their agentic AI capabilities, expanding multilingual and industry-specific offerings, and strengthening their engineering team. The round remains open and may see additional participation.
For the agentic ecosystem, Gnani represents the voice layer that most agent frameworks still lack. We've seen Voxtral TTS (open-source text-to-speech), Gemini Flash Live (native voice tool use), and Cohere Transcribe (open-source ASR). Gnani completes the picture for emerging markets — a full-stack voice agent platform that doesn't route everything through US-based APIs. In a world where agents need to talk to customers in Hindi, Tamil, Bengali, and Bahasa, that matters.
https://www.businesstoday.in/technology/story/gnaniai-raises-10-million-in-funding-from-aavishkaar-capital-to-scale-global-voice-ai-push-523218-2026-03-31
← Back to all articles
Gnani.ai, a Bengaluru-based voice-first agentic AI startup, raised $10 million in its ongoing Series B round led by Aavishkaar Capital, with existing investor Info Edge Ventures also participating. The announcement dropped on March 31, 2026.
Here's what makes Gnani different from the crowd of voice AI startups. They built the entire stack in-house — speech recognition, voice synthesis, and a proprietary 14 billion-parameter speech-to-speech model. No dependency on OpenAI Whisper or third-party TTS. This is sovereign AI infrastructure, designed to work across 12+ languages with the kind of latency and accuracy that enterprise clients in banking, healthcare, and insurance demand.
The scale is real. 30 million voice interactions processed daily across 200+ enterprise clients. Founded in 2016 by Gopalan and Ananth Nagaraj, this isn't a startup that raised on a pitch deck — they've been grinding for a decade.
The funding will go toward global expansion, advancing their agentic AI capabilities, expanding multilingual and industry-specific offerings, and strengthening their engineering team. The round remains open and may see additional participation.
For the agentic ecosystem, Gnani represents the voice layer that most agent frameworks still lack. We've seen Voxtral TTS (open-source text-to-speech), Gemini Flash Live (native voice tool use), and Cohere Transcribe (open-source ASR). Gnani completes the picture for emerging markets — a full-stack voice agent platform that doesn't route everything through US-based APIs. In a world where agents need to talk to customers in Hindi, Tamil, Bengali, and Bahasa, that matters.
https://www.businesstoday.in/technology/story/gnaniai-raises-10-million-in-funding-from-aavishkaar-capital-to-scale-global-voice-ai-push-523218-2026-03-31
Comments