Dubformer, an Amsterdam-based startup specializing in professional AI dubbing, announced a $3.6 million seed funding round to push the boundaries of emotional expression in AI dubbing, enhancing media localization for TV, broadcast, and film productions.
The round was led by Almaz Capital, with s16vc and FinSight as key investors, along with angel investors such as Arul Menezes, founder of Microsoft Translator, and Funa Maduka, former
head of International Original Film at Netflix. This funding will enable Dubformer to advance its proprietary AI technologies and enhance emotional depth in voice performance, aligning with its mission to make content accessible and enjoyable globally.
Market problems
Despite impressive advances in the AI dubbing industry, the quality of AI-generated dubbed content still falls short of being convincingly realistic. Recent research indicates that emotional voices are likely to be identified as human, while neutral tones are generally associated with AI.
This distinction not only affects how viewers perceive the content but also influences their overall enjoyment.
The video localization market, currently worth $7 billion, is growing at a compound annual growth (CAGR) of 16% to meet the increasing global demand. In addition to independent companies offering dubbing solutions, major players also invest in internal innovation to enhance content localization and expand audience reach.
Dubformer’s innovation
Dubformer addresses the key market challenges with Emotion Transfer, a technology which enhances dubbing by transferring emotions instead of cloning voices, as conventional AI dubbing solutions do.
At the core of Emotion Transfer lies its ability to generate fine-grained local aspects of speech — such as intonation, emotion, and pace — using source audio as a reference. This is
generative dubbing, not mere voice replication. Because languages have unique prosodic features, perfectly mimicking all speech nuances is impossible. This approach delivers dubbing
that is significantly more natural and immersive compared to conventional technologies such as text-to-speech or voice cloning.
“With Emotion Transfer, we also capture the acoustic characteristics — such as echoes and reverberations — enhancing the overall experience. Our focus is on creating full immersion,
making sure the dubbing never feels out of place,” said Anton Dvorkovich, Founder & CEO of Dubformer.
In 2024, Dubformer’s AI dubbing technology competed in WMT, the leading annual machine translation competition held as part of the conference of the same name. WMT attracts major
players from both industry and academia, including Google Translate, DeepL, ChatGPT, and Claude. Dubformer’s solution ranked highest for speech translation, and highest overall in
multiple languages, including Spanish.
Background and plans
Dubformer was founded in September 2023 by Anton Dvorkovich, a machine translation expert with 10+ years of experience and former Head of NLP & Speech at Yandex. The core team, experts in AI and speech technology, previously launched the world’s largest AI dubbing solution which had over 5M users and 150M minutes watched daily.
Since its launch, Dubformer has grown rapidly, gaining traction among media companies and content creators. Over the past six months, the platform’s growth has accelerated significantly, averaging more than 50% month-over-month, reflecting the surging demand for high-quality, emotionally expressive AI dubbing.
Dubformer now serves over 200 clients across the U.S. and Europe, including Paramount, Little Dot Studios, and Mainstream Media. It also has partnered with companies such as Voxx (Los
Angeles), Presto (Czech Republic), and Gobavo (Turkey).
“In a rapidly evolving AI dubbing landscape, Dubformer is committed to driving the industry forward by addressing its key challenges. Through capturing the emotional essence of content,
we enable media companies to expand their global audience reach while reducing localization costs — all without compromising viewer satisfaction,” said Anton Dvorkovich, Founder & CEO
of Dubformer.
Dubformer’s technology is also accessible to users and creators through its open platform and
API.
