Berlin-based startup Mirelo has secured a massive $41 million investment to solve the biggest problem in generative video.
The company aims to become the definitive audio layer for visual content creators worldwide. This significant capital injection was led by venture capital titans Index Ventures and Andreessen Horowitz (a16z). The funding round signals a major shift in investor interest toward the lagging audio sector of the generative AI boom.
The End of the Silent AI Era
The generative AI revolution has moved at breakneck speed for text and video. Tools like ChatGPT and various video generators have transformed creative workflows. Yet sound has remained a stubborn bottleneck. Most AI-generated videos today resemble the silent films of the 1920s. They lack synchronized, realistic audio that matches the on-screen action. Mirelo intends to fix this disconnect permanently.
Mirelo has developed proprietary AI models that generate synchronized sound for video in seconds.
The startup was founded in 2023 by CJ Simon-Gabriel and Florian Wenzel. Both founders are former musicians who met while working as AI researchers at Amazon. Their unique blend of artistic background and engineering prowess sits at the core of the company’s mission.
CJ Simon-Gabriel, the CEO of Mirelo, draws a parallel to cinema history. He noted that video without sound lacks feeling and atmosphere. He compares the current state of AI video to the era of silent films.
“Mirelo’s first step is about democratising access, empowering everyone to create the sound that their videos deserve,” Simon-Gabriel said.
The founders believe that creators should not be bogged down by technical limitations. Their goal is to let professionals focus on expression rather than the tedious mechanics of synchronization.
Mirelo AI sound generation technology visualization audio wave
Under the Hood of Mirelo SFX v1.5
Creating realistic audio for video is notoriously difficult. It requires precise timing and acoustic understanding. Editors currently spend hours combing through stock libraries to find the right door slam or footstep. They then have to manually sync these sounds to the video timeline. This process is expensive and kills creativity.
Mirelo recently released its new video-to-sound model called Mirelo SFX v1.5. This technology automates the entire sound design process. A user simply uploads a video. The model analyzes the pixels and motion. It then produces a matching audio track for everything happening on screen.
The startup claims its models require 50 times less computing power than typical Large Language Models.
This efficiency is a critical advantage. Lower compute requirements mean the tool can run faster and cheaper than competitors. Mirelo says its tech can generate various soundtrack versions faster than real-time. This speed is essential for modern content creators who work on tight deadlines.
The technology is built from scratch. Mirelo trains its models on data for which it has secured specific licensing deals. This approach avoids the copyright pitfalls that have plagued other AI companies.
A Powerhouse of European and US Backing
Raising $41 million in a seed round is incredibly rare. It highlights the immense potential investors see in Mirelo’s technology. The round included participation from Berlin-based investor Atlantic and California-based TriplePoint Capital.
The startup also attracted angel investment from notable tech figures. These include Arthur Mensch, the co-founder and CEO of Mistral, and Antoine Le Nel from Revolut. This support from European tech luminaries suggests strong confidence in the Berlin ecosystem.
Guido Appenzeller, a partner at Andreessen Horowitz, explained the firm’s bullish stance. He noted that Mirelo is tackling one of the least explored areas of generative media.
“CJ and Florian have assembled a research-driven team whose breakthroughs in tokenisation, data curation, and conditioning rival far larger efforts,” Appenzeller said.
He added that a16z is excited to back Mirelo as they scale their technology for the next generation of video models. The firm views specialized sound effect creation as a technically challenging frontier.
Transforming Gaming and Social Media
Mirelo has a vision that extends far beyond simple video clips. The company wants to integrate its API into gaming worlds and social media platforms.
Video games require dynamic audio that reacts to player actions. Mirelo’s tech could generate these sound effects on the fly. This would reduce game file sizes and create infinite audio variations.
“Our bigger mission is to become the audio layer for all visual content across videos, gaming, social media, films and beyond,” Simon-Gabriel stated.
The startup currently operates with a lean team of 10 people. They plan to use the new funds to advance their technology and grow their customer base. Customers currently include individual creators and small studios. However, the API offering signals a push toward enterprise clients who want to embed audio generation into their own tools.
Florian Wenzel highlighted the connection between their team’s background and their product.
“There’s a deep affinity between music and engineering,” Wenzel said. “Maybe that’s why so many of Mirelo’s team are musicians, and why musicians have always been early adopters of new technology.”
He believes the intersection of mathematical precision and expressiveness draws people to both fields. This philosophy drives the product development at Mirelo. They are not just building a utility. They are building an instrument for the AI age.
The funding brings Mirelo’s total capital raised to approximately $44 million. With a significant war chest and a product that addresses a glaring gap in the market, Mirelo is poised to define how we hear the internet in the coming years.
