Audio engineers often struggle with recordings tainted by background noise, but a German firm called AI-coustics is trying to change this by using generative AI in a novel way to improve speech clarity in movies.
AI-coustics made its debut with funding of €1.9 million, offering a technology that goes beyond traditional noise suppression. According to Fabian Seipel, co-founder and CEO, their technology is made to improve audio quality across all devices and speakers.
Their goal is to make all digital interactions as clear as studio broadcasts, whether it is a conference call or a social media video.
AI-coustics was established in 2021 as a result of the mutual dissatisfaction of audio engineer Seipel and machine learning lecturer at the Technical University of Berlin Corvin Jaedicke, about the poor audio quality of online educational content. Seipel's own struggles with hearing loss from his early work in music production are the inspiration behind his personal mission to enhance digital audio clarity.
AI-coustics sets itself apart in the crowded field of AI-enabled voice improvement products with its creative approach to noise reduction technology development.
The business's algorithm, which rewards users for adding to their training dataset, is based on speech samples that were captured in their Berlin studio. They then use this dataset to improve their noise-reduction AI, which during its training process accurately mimics a variety of audio flaws.
Concerns exist, though, regarding the contributors' one-time payment approach and possible speech recognition biases. In order to counteract prejudices, AI-coustics is dedicated to diversifying its contributor base in order to guarantee the efficacy of the technology in all linguistic contexts and demographic contexts.
Three different video clips were run via AI-coustics' platform to evaluate the efficacy of the technology. The results showed a significant improvement in speech clarity by lowering background noise.
AI-coustics hopes to use their technique for both pre-recorded and real-time speech augmentation. It may even incorporate its technology into consumer goods to automatically boost voice quality. At the moment, the startup provides an SDK, web application, and API to make audio and video post-processing easier.
With a combination of on-demand, subscription-based, and license income streams, AI-coustics has drawn both a large user base and enterprise clients. Using its recent funding and support system, the company intends to grow its workforce and improve its speech-enhancement strategy in order to meet market expectations and negotiate the venture capital ecosystem.
In response to worries that artificial intelligence (AI) could replace workers, Seipel emphasizes how AI-coustics may expedite the audio production process, freeing up human engineers for other duties while guaranteeing high-quality voice output. The startup's technology promises broad applicability across devices and content, aiming to solve the widespread problem of poor audio quality in digital communication.