At IBC2026, LALAL.AI shows its stem splitter running fully offline inside a device, alongside new dialogue-isolation demos aimed at post-production, dubbing, and localization workflows.

AMSTERDAM, NL / ACCESS Newswire / September 11, 2026 / LALAL.AI, the Webby-winning AI-powered audio processing company, unveils its first embedded hardware integration at IBC2026, running its stem-separation engine natively inside Neko, a compact wireless AI mini studio for guitar players from hardware partner Neko Engineering. For LALAL.AI, the integration means a shift from browser-based tools and studio plugins into on-device AI, with separation running fully offline in real time, with no cloud round-trip and no per-inference cost.

IBC visitors to Stand 14.C45 in Hall 14 (Future Tech) will be able to see the Neko device separate a track into vocals, drums, bass, and guitar live, entirely on-device, without an internet connection.

“For six years our tech has lived in a browser tab. Putting it inside a physical device, running in real time with nothing sent to a server, is a different kind of product decision. It’s a signal that separation-quality AI doesn’t have to depend on the cloud to be useful,” said Nik Pogorsky, LALAL.AI, co-founder.

“We wanted stem separation that a musician could rely on mid-jam, not something that waits on a network connection. LALAL.AI’s engine was the one that held up in real time, on-device hardware, without cutting corners on separation quality,” said Vadim Smirnov, CEO, Neko Engineering.

LALAL.AI solves real production problems, live at the stand

Alongside the Neko integration, LALAL.AI will also demonstrate its technology across a series of production scenarios drawn directly from broadcast, post-production, dubbing and localization workflows:

  • Field audio cleanup, turning noisy, wind- and traffic-affected location recordings into clean dialogue in minutes without a re-record.
  • Dialogue isolation for dubbing without a delivered M&E track, recovering clean dialogue from a finished mix so a new-language dub can be laid in over the original background, without re-recording the entire soundtrack.
  • Clearing music beds that trigger Content ID blocks. LALAL.AI Voice Cleaner will demonstrate the ability to remove a stadium or background music bed from a clip while preserving commentary and ambience, so social and highlight content can clear rights checks.
  • Stem extraction from a final mix with no multitrack available, pulling vocal and instrumental stems out of a delivered mix as if returning to the original session.
  • Removing room reverberation from speech recorded in untreated spaces, such as conference halls and lecture venues.
  • On-premises processing via Lyra-powered LALAL.AI’s VST3/AU plugin, useful for studios and broadcasters with strict data-security and NDA requirements. Running inside DAWs and NLEs including Ableton Live, Reaper, DaVinci Resolve (Fairlight) and Adobe Premiere Pro without material leaving the local machine.

“The Neko integration and the demos on our stand are really the same story told twice. Whether it’s a guitarist splitting a track in real time or a dubbing studio pulling clean dialogue out of a mix with no delivered M&E, the problem is the same: getting usable, controllable audio out of something that wasn’t handed to you clean,” said Anastasia Shikina, LALAL.AI Business Development Manager, who’ll be at the IBC stand.

Availability

LALAL.AI’s Voice Cleaner, Stem Splitter, Voice Cloner, VST plugin, and related tools are available now at lalal.ai.

The Neko device, with LALAL.AI-powered Stem Splitter built in, is available for pre-order at https://neko.engineering/promo for $349, a reduction from its $499 retail price, with shipping to backers beginning in 2027.

Visit LALAL.AI at IBC2026

LALAL.AI will be on Stand 14.C45, Hall 14 (Future Tech), RAI Amsterdam, September 11-14, 2026. To book a meeting on-site, please contact Anastasia Shikina, LALAL.AI’s Business Development Manager, via Calendly or LinkedIn.

About LALAL.AI

LALAL.AI is a Webby-winning audio processing platform trusted by nearly 7 million sound professionals worldwide. Its AI-driven audio separation and voice modification solutions for the global media, entertainment, and technology sectors are powered by proprietary deep-learning algorithms. Initially recognized for its high-performance cloud tools and studio plugins, LALAL.AI’s ecosystem now spans scalable enterprise APIs and real-time embedded hardware integrations. By eliminating technical barriers in dialogue isolation, localization, and audio asset management, LALAL.AI enables global media workflows and hardware developers to optimize audio quality and accelerate production pipelines with ultimate data security and zero cloud dependency.

To learn more, visit www.lalal.ai.

About Neko Engineering

Neko Engineering is a UAE-based company building Neko, a wireless pocket studio designed to change how people play guitar, so that anyone, regardless of skill level, can become a musician.

Neko brings everything to play guitar-great tone, wireless connectivity, a DAW, stem separation, tab transcription, and more into a single portable device that works fully offline.

Built on the latest audio hardware and AI, Neko’s design won the Red Dot Award: Design Concept 2026. Founded by Vadim Smirnov, previously CPO of the most popular Arabic-language AI assistant and speakers in the UAE. The company launches Neko on Kickstarter in October 2026.

To learn more, visit www.neko.engineering.

Media Contact

Klara Alexeeva
LALAL.AI Communications Manager
klara.alexeeva@lalal.ai

SOURCE: LALAL.AI

View the original press release on ACCESS Newswire

About The Author
LALAL.AI