Modulate, a Boston-based company that builds audio-native artificial intelligence, has raised $25 million in a round led by Future Ventures, with participation from Hyperplane and Lakestar.
Its platform, Velma, does not rely on a single large model. Instead it runs an array of more than 100 small models that listen to speech and return signals about it: transcription, analysis of tone, emotion and intent, and detection of synthetic audio such as deepfakes and AI-generated music. The company sells the resulting analysis software to enterprises for uses including fraud prevention, deepfake detection and supervision of AI agents.
The new capital is intended to accelerate development of those audio-native models and to expand Modulate's developer and partner ecosystem. The round brings the company's total funding to around $60 million, according to media reports; it had previously raised $41 million at a $170 million valuation.
The raise lands at a moment when platforms and regulators are under pressure to detect AI-generated voice content — a problem that has fuelled impersonation scams and disputes over consent in synthetic media. Modulate's bet is that audio analysis, understanding not just what was said but whether the speaker is human, becomes an enterprise requirement rather than a specialty tool.
