The perpetual struggle with noisy audio recordings during interviews and speeches has long vexed audio engineers worldwide. However, a pioneering German startup aims to revolutionize this challenge by leveraging generative AI to refine the clarity of voices in videos.
Emerging from stealth mode with €1.9 million in funding, AI-coustics introduces a groundbreaking solution beyond conventional noise suppression, tailored to function seamlessly across various devices and speakers.
Fabian Seipel, co-founder and CEO of AI-coustics, envisions a future where every digital interaction, from conference calls to social media videos, rivals the clarity of professional studio broadcasts.
Seipel, a trained audio engineer, joined forces with Corvin Jaedicke, a machine learning lecturer at the Technical University of Berlin, in 2021 to establish AI-coustics. Their motivation stemmed from encountering subpar audio quality in online courses during their time at TU Berlin.
Driven by a personal mission to combat poor audio quality in digital communications, Seipel emphasizes the transformative potential of their technology in enhancing speech quality and intelligibility.
While existing AI-powered noise-suppression solutions abound, AI-coustics distinguishes itself through a unique approach to developing AI mechanisms for noise reduction.
The startup employs a model trained on speech samples meticulously recorded in its Berlin-based studio. Contributors are compensated for recording samples, which are then integrated into the dataset to refine AI-coustics’ noise-reduction model.
Seipel acknowledges potential concerns regarding the compensation model for contributors and addresses the paramount issue of bias mitigation in AI algorithms. AI-coustics prioritizes recruiting a diverse pool of speech sample contributors to ensure inclusivity and mitigate biases.
Initial tests on AI-coustics’ platform demonstrate its efficacy in enhancing audio clarity, affirming its potential for real-time speech enhancement and integration into various devices.
With revenue streams from subscriptions, on-demand pricing, and licensing, AI-coustics has garnered five enterprise customers and boasts 20,000 users. Future plans include team expansion and continuous refinement of the speech-enhancing model.
Seipel envisions AI-coustics as a complementary tool rather than a job usurper, offering content creators and broadcast managers a means to automate audio production tasks efficiently while upholding speech quality standards.
The startup secured funding from Connect Ventures, Inovia Capital, FOV Ventures, and Ableton CFO Jan Bohl, positioning itself for further growth and innovation in the audio technology landscape.
