Modulate raises $25M for native AI-based frontier audio
- ModulateCompany
Future VenturesInvestor
HyperplaneInvestor
LakestarInvestor
Modulate raised $25 million in a venture round led by Future Ventures, with Hyperplane and Lakestar participating, to accelerate its audio‑native AI platform and developer ecosystem.
Modulate has raised $25 million in a venture funding round led by Future Ventures, with participation from Hyperplane and Lakestar, on Sept. 28, 2026. The capital brings the company’s total financing to $60 million and is earmarked for scaling its audio‑native AI models and expanding its developer and partner ecosystem.
Deal Terms
The round was not disclosed as a priced equity round, and valuation multiples were not disclosed. Future Ventures acted as the lead investor, while Hyperplane and Lakestar joined as co‑investors. The funding will be allocated across AI/ML research, product engineering, developer relations, and partnership initiatives.
Strategic Rationale
Modulate’s flagship Velma platform processes more than 600 million hours of audio and currently ranks #1 on Hugging Face benchmarks for both transcription and deep‑fake detection. By analyzing audio directly, Velma delivers up to 2× higher accuracy and 7× fewer false positives than traditional large language models, while consuming up to 1,000× less compute. The new capital enables the company to broaden its API catalog, launch additional SDKs, and deepen integrations with security, customer‑experience, and voice‑AI vendors. As voice becomes a primary interface for AI, Modulate’s audio‑native approach addresses a gap that transcript‑only solutions cannot solve, positioning the firm as a foundational layer in the emerging voice‑AI stack.
The infusion also supports hiring across research, engineering, and go‑to‑market teams, reinforcing Modulate’s push into high‑growth verticals such as fraud prevention, trust‑and‑safety, and enterprise communications. With its models already deployed in healthcare, gaming, and social platforms, the company is poised to capture a larger share of the expanding market for real‑time audio intelligence.
Overall, the round underscores investor confidence in Modulate’s differentiated technology and its potential to become the de‑facto infrastructure for audio‑centric AI applications.
Why It Matters
For Modulate, the $25 million infusion accelerates product rollout and deepens its foothold in high‑value verticals where real‑time audio analysis is mission‑critical. Competitors that rely on transcript‑first pipelines will face pressure to match Modulate’s accuracy and efficiency, especially as enterprises demand lower compute costs for large‑scale voice deployments. The expanded developer tooling also raises the barrier to entry for new entrants, as partners can integrate sophisticated audio signals without building bespoke models. Existing voice‑AI platforms may need to augment their stacks with Modulate’s APIs to stay competitive in fraud detection, empathy modeling, and safety monitoring.
From an investor perspective, Future Ventures, Hyperplane, and Lakestar are betting on a niche yet scalable SaaS model where recurring revenue can be generated through API usage fees (e.g., $0.03 per hour for transcription). The round validates the market’s appetite for specialized AI layers that sit beneath broader LLM offerings, potentially prompting follow‑on funding for other audio‑focused AI startups.
The deal also signals a shift in the AI infrastructure landscape: as large language models dominate headline coverage, specialized modalities like audio are emerging as critical differentiators. Companies that can monetize these capabilities through a SaaS pricing model stand to capture high‑margin, recurring revenue streams as voice interfaces proliferate across consumer and enterprise products.
Key Points
- Modulate secured $25 million in venture funding, bringing total capital to $60 million.
- Future Ventures led the round; Hyperplane and Lakestar participated as co‑investors.
- Funding will accelerate development of audio‑native AI models and expand the developer ecosystem.
- Velma platform has processed over 600 million hours of audio and ranks #1 on Hugging Face benchmarks for transcription and deep‑fake detection.
- Modulate’s technology is deployed across fraud prevention, voice‑AI supervision, customer experience, and trust‑and‑safety use cases.
Analysis
While the round’s valuation was not disclosed, the $25 million raise highlights the premium investors are placing on modality‑specific AI infrastructure. Modulate’s SaaS model—charging per‑hour usage for transcription and deep‑fake detection—offers a clear path to recurring revenue and high gross margins, traits that align with investor expectations for scalable AI businesses. The company’s ability to process audio at 1,000× lower compute cost than monolithic models creates a defensible cost advantage, enabling price‑competitive API offerings that can drive strong net‑revenue retention as developers embed the service into mission‑critical workflows.
The broader AI market is witnessing a fragmentation into specialized layers (vision, audio, code) that complement large language models. Modulate’s focus on audio‑native intelligence positions it to capture a growing slice of the voice‑AI market, which analysts project to expand at double‑digit CAGR as voice interfaces become ubiquitous in consumer apps, enterprise communications, and security solutions. For SaaS operators, the raise underscores the importance of building domain‑specific APIs that can be monetized via usage‑based pricing, while for investors it reinforces the thesis that vertical AI SaaS companies can achieve attractive multiples by delivering high‑value, low‑compute services. As more enterprises adopt voice‑first experiences, Modulate’s platform could become a core component of the AI stack, driving both top‑line growth and higher valuation benchmarks for audio‑centric SaaS firms.