Deepgram hits $1.3B valuation with $130M Series C for Voice AI
Deepgram, a leading Voice AI and speech-to-text platform, has raised a substantial $130 million Series C funding round, pushing the company to a $1.3 billion valuation. The fresh capital underscores surging demand for real-time AI agents, enterprise-grade voice analytics, and automated workflows across industries, from contact centers to restaurants.
The round is backed by major investors including AVP (Allianz Ventures Partners), Twilio, and other prominent venture firms, signaling strong institutional confidence in the company’s technology and long-term market potential.
Strategic acquisition of OfOne to expand into restaurant automation
As part of its growth strategy, Deepgram has acquired OfOne, a specialist in restaurant automation and conversational ordering systems. OfOne’s technology focuses on using AI agents to handle real-time voice interactions for quick-service and fast-casual restaurants, helping brands reduce wait times, cut labor costs, and improve order accuracy.
By integrating OfOne’s stack, Deepgram aims to deliver end-to-end voice automation solutions that go beyond transcription. The combined offering can power intelligent drive-thru systems, phone-based ordering, and voice-enabled kiosks, all built on high-accuracy speech recognition and natural language understanding.
For the restaurant sector, which continues to face tight margins and chronic staffing challenges, this move positions Deepgram as a key infrastructure provider for next-generation, AI-driven customer experiences.
New San Francisco hub to anchor AI talent and partnerships
Alongside the funding and acquisition, Deepgram is launching a new hub in San Francisco, one of the world’s most competitive markets for AI research and engineering talent. The hub will serve as a center for product development, customer collaboration, and go-to-market operations.
By establishing a strong presence in the Bay Area, Deepgram is positioning itself closer to leading AI startups, large enterprise customers, and strategic partners that are actively deploying real-time voice interfaces in their products and services. The San Francisco office is expected to accelerate hiring across machine learning, infrastructure engineering, and enterprise sales.
Scaling real-time Voice AI for agents and enterprises
Deepgram has built its reputation on delivering highly accurate, low-latency speech-to-text APIs that can process large volumes of audio in real time. This capability is critical for AI agents that must listen, understand, and respond instantly during live calls, support sessions, or in-person interactions.
Powering AI agents and virtual assistants
The new funding will allow Deepgram to enhance its core AI models, support more languages and accents, and improve robustness in noisy environments. These enhancements are particularly important for:
- Contact centers deploying AI agents to handle inbound and outbound calls.
- Enterprises building virtual assistants and voicebots for customer support.
- Developers integrating real-time transcription into collaboration tools, meeting platforms, and analytics systems.
By focusing on speed and accuracy, Deepgram aims to become a foundational layer for the emerging ecosystem of voice-first applications.
From transcription to full-stack voice intelligence
While speech-to-text remains the company’s core capability, Deepgram has increasingly moved up the stack into voice intelligence and conversation analytics. Its platform can extract insights such as sentiment, intent, and key topics from customer interactions, enabling businesses to measure performance and automate quality assurance.
The acquisition of OfOne accelerates this shift from being a general-purpose transcription engine to a provider of domain-specific, workflow-ready voice AI solutions. In restaurants, that translates into automated order taking. In call centers, it powers agent assistance, call summarization, and compliance monitoring. In software products, it underpins voice-enabled interfaces that feel natural and responsive.
Investor backing highlights confidence in Voice AI market
Support from investors such as AVP and Twilio reflects a broader conviction that voice will be a dominant interface for both consumers and enterprises. As companies race to build AI-native products, demand for reliable, scalable Voice AI infrastructure is surging.
Twilio, known for its communications APIs, has long been a bellwether for developer-focused platforms. Its involvement signals strong alignment between programmable communications and advanced voice recognition, particularly as customers seek integrated solutions that combine telephony, messaging, and AI automation.
For AVP and other institutional backers, the deal represents a bet that Deepgram can capture a significant share of the rapidly expanding Voice AI market, spanning sectors such as financial services, healthcare, retail, logistics, and hospitality.
What the Series C means for the Voice AI landscape
The $130 million Series C provides Deepgram with the resources to scale infrastructure, invest heavily in AI research, and deepen its product portfolio around real-time agents and vertical-specific solutions like restaurant automation. It also intensifies competition among speech technology providers racing to offer lower latency, higher accuracy, and more flexible deployment models.
As enterprises look beyond generic chatbots toward more capable, multimodal AI agents that can listen, speak, and act, platforms like Deepgram are likely to play a central role. The company’s push into San Francisco, its acquisition of OfOne, and its strengthened balance sheet collectively signal an ambition to become one of the default choices for developers and businesses building the next generation of voice-driven experiences.
For customers, the developments promise faster innovation cycles, better-performing voice interfaces, and increasingly automated workflows that can operate reliably at scale and in real time.