AudAi launches zero-shot AI voice editing solution 'Gluent'
The AMW Read
Incremental update to the AI voice editing space; Gluent is a new product from an established Korean startup but does not fundamentally shift the competitive landscape.
AudAi launches zero-shot AI voice editing solution 'Gluent'
South Korean voice AI startup AudAi (오드아이) has officially launched Gluent (글루언트), a zero-shot speech-to-speech (STS) editing solution designed for professional creators and audio content studios. Gluent allows users to edit specific words in recorded speech by simply modifying text, without requiring separate data training — preserving the original speaker's tone, rhythm, and spatial ambiance. The company claims the tool can reduce re-recording costs by over 70%. Current features include pronunciation correction and multilingual voice conversion; a recording-environment enhancement feature is planned. Gluent is available under a subscription model with Free, Basic, Creator, and Business tiers.
Why it matters: This launch targets a persistent inefficiency in AI voice: while short-form voice synthesis is commoditized, long-form content (audiobooks, dubbing, film/TV) still suffers from high production timelines and inconsistent voice consistency across hours of audio. Gluent addresses the context-engineering moat problem in audio post-production — the challenge of maintaining coherent prosody and timbre across time spans where traditional TTS or voice cloning breaks down. The zero-shot approach removes the training overhead that has limited STS adoption in professional workflows. AudAi's selection for the AWS Jungle accelerator program (part of South Korea's Global Corporate Collaboration Program) signals intent to distribute via hyperscaler channels, a familiar pattern in the AI tooling layer: use cloud marketplace distribution to reach global creative studios.
Grounded expert take: AudAi has prior credibility from high-profile collaborations — including AI vocal work with K-pop label SM Entertainment's virtual artist nævis, and AI-enhanced live performance with soprano Sumi Jo — giving it a meaningful track record in Korean entertainment AI. The subscription-pricing structure across hobbyist, pro, and studio tiers mirrors the tiered-access model that has worked for AI music and video editing tools. The open question is whether AudAi can differentiate Gluent enough from incumbents like Sonantic (acquired by Spotify), Respeecher, and ElevenLabs, which already offer speech editing but focus on different workflows. The answer likely depends on Gluent's long-form consistency and AWS marketplace adoption velocity.
