Modulate Raises $25M to Expand Audio AI for Voice Gaming and Safety

Modulate, an audio-native AI company founded by MIT alumni, has raised $25 million in new funding to expand its voice analysis technology. Future Ventures led the round, with participation from Hyperplane and Lakestar. This brings Modulate's total funding to $60 million.

The company's flagship Velma platform analyzes audio directly to identify emotion, tone, intent, synthetic speech, and conversational behaviors. Unlike traditional large language models that work from transcripts alone, Velma processes raw audio to detect fraud attempts, deepfake voices, AI agent failures, harassment, and policy violations. The technology operates in real time, allowing applications to intervene during conversations rather than after the fact.

Modulate's underlying Ensemble Listening Model (ELM) architecture uses more than 100 specialized audio models rather than relying on a single massive foundation model. This approach has demonstrated up to 1,000 times greater efficiency than single large model solutions, significantly reducing the compute, cost, and energy required to analyze audio at scale.

The company's technology has already proven itself at scale. Modulate's models now analyze more than 10 million hours of audio each month and have processed over 600 million hours in total. Its transcription technology and deepfake detection both rank first on Hugging Face's public benchmarks. The deepfake detection achieves 98.9% accuracy on public benchmark data, while the transcription API is priced at $0.03 per hour for batch processing.

Current applications span fraud prevention, voice AI agent supervision, customer experience, and trust and safety. Organizations use Modulate to protect healthcare institutions from deepfake attacks, enhance voice AI agents' emotional recognition and empathy, reduce extremism and harassment on social platforms, monitor voice agent performance, detect child grooming in voice conversations, and protect agents in high-risk scenarios through advanced voice masking.

Carter Huffman, CEO and cofounder of Modulate, said in a statement: "Voice is becoming a primary interface for AI, and that creates a whole new set of problems that can't be solved from a transcript. We're already using audio-native AI to protect organizations from deepfake attacks, help voice agents understand emotion and respond with more empathy, identify dangerous behavior in online conversations, and monitor whether voice agents are actually performing the way they're supposed to."

The new capital will support expansion across AI and machine learning research, product and engineering, developer relations, and partnerships. Modulate plans to broaden its APIs, models, and deployment options while building new industry-specific models, expanding developer tooling with new SDKs and APIs, and creating partner integrations.

Steve Jurvetson, cofounder of Future Ventures and board member of SpaceX AI, said: "Modulate has gained a significant technical lead in audio-native AI, and the market opportunity is expanding quickly. The team has proven these models in some of the most demanding voice environments in the world, and we're now seeing the need for that technology to expand well beyond where it started into AI agents, security, customer experience, and more."

Huffman added: "Developers shouldn't have to rebuild the audio intelligence layer every time they create a new voice experience. Our mission is to build the models and infrastructure that let them focus on the application they want to create."