Modulate has successfully raised $25 million to advance its audio-native AI platform, focusing on tackling deepfake voice threats that challenge modern enterprises. This substantial funding round, led by Future Ventures and supported by Hyperplane and Lakestar, elevates the company’s total investment to $60 million.
The fresh capital is set to boost Modulate’s AI research, engineering, and development efforts, as well as expand its partnerships and deployment options. The company’s audio analysis technology, which goes beyond mere transcript examination, evaluates aspects like emotion, tone, and intent to identify synthetic speech.
Advancements in Audio AI Technology
At the core of Modulate’s innovations is Velma, a platform designed to understand conversations in real-time by detecting fraud, harassment, and policy violations. Its Ensemble Listening Model (ELM) employs over 100 specialized audio models to enhance inference efficiency while minimizing resource consumption.
Deepfake detection is a critical feature, with Modulate achieving a 98.9% accuracy rate on the Speech DF Arena benchmark. The platform’s model can deliver results after just 2.5 seconds of speech, although Modulate advises clients to customize detection thresholds based on their specific needs to avoid false positives.
Impact on Various Industries
Modulate’s technology processes over 10 million hours of audio monthly, reaching the top of the Hugging Face’s Open ASR Leaderboard for transcription. Their API services are economically priced, making them accessible across various sectors.
The funding will help Modulate expand APIs, develop industry-specific models, and enhance integration capabilities. The company’s focus spans fraud prevention, healthcare security, gaming moderation, and more, aiming to provide real-time solutions to preemptively tackle threats.
Future Outlook for Voice AI
As audio becomes a dominant AI interface, Modulate’s CEO, Carter Huffman, emphasizes the need for robust voice analysis technology. The company sees audio understanding as crucial for authenticating callers and monitoring automated agents.
With advancements in synthetic voice technology, combining audio AI with multifactor authentication and human oversight remains essential. Modulate’s innovations aim to ensure that audio AI supports security teams without becoming a sole point of failure.
This latest funding round underscores Modulate’s commitment to leading the charge against audio-based threats, ensuring secure and reliable communications in an increasingly digital world.
