Mirelo secures $41M from Index and a16z to address the silent-audio challenge in AI video.

Mirelo’s $41M funding round from Index and a16z for AI video and audio technology. Mirelo, a Berlin-based AI startup, has raised $41 million in seed funding led by Index Ventures and Andreessen Horowitz to tackle a glaring gap in generative AI: most AI video creation tools produce stunning visuals but output mute files. The two-year-old company, founded by AI researchers and musicians CJ Simon-Gabriel and Florian Wenzel who met at Amazon, released Mirelo SFX v1.5 earlier this year—an AI model that interprets videos to add synchronized sound effects. With a freemium model that includes a recommended plan for creators priced at €20 per month (approximately $23.50), the startup targets amateurs and prosumers hoping to unmute AI-generated videos. The funding brings Mirelo’s total capital raised to $44 million, including a pre-seed round led by Berlin-based firm Atlantic, and includes angel backing from notable tech figures like Mistral CEO Arthur Mensch and Hugging Face chief science officer Thomas Wolf.

While in stealth mode and resource-constrained, Mirelo watched as large competitors including Sony, Tencent, China’s Kuaishou-owned Kling AI, and a16z-backed ElevenLabs released video-to-sound effects models. Even Google’s Gemini video generator now incorporates soundtracks powered by DeepMind’s Veo 3.1 video-to-audio model. Simon-Gabriel acknowledges the competition but sounds vindicated, comparing the situation to silent movies versus talkies and noting that people are suddenly realizing they should add sound to AI videos. Mirelo differentiates itself through its narrow focus on sound effects rather than full-stack audio solutions, with models that require 50 times less computing power than typical large language models while delivering faster-than-real-time generation. The company’s 10-person team is expected to double or triple by next year to accelerate research and development alongside go-to-market strategy.

Simon-Gabriel emphasized the critical importance of audio, quoting George Lucas’s assertion that sound is 50% of the movie-going experience, adding it’s actually an understatement since the same images can create completely different atmospheres depending on the soundtrack. To address generative AI’s contentious copyright issues, Georgia Stevenson from Index Ventures noted that Mirelo bases its models on public and purchased sound libraries and is signing revenue-sharing partnerships that respect artists’ rights. The startup has AI music generation on its roadmap, though it currently focuses primarily on sound effects, with both co-founders being musicians themselves. Mirelo plans to expand into interactive environments like adaptive game worlds or dynamic augmented reality spaces where real-time sound generation responds to player behavior, positioning itself as the audio layer for all visual content across videos, gaming, social media, films, and beyond in an era where creating professional-grade audio for AI-generated content remains largely manual and time-consuming.