Supertone, the AI Voice Company owned by K-Pop Giant Hybe, has introduced a new AI Text-to-Voice tool that claims to generate “high-quality, expressive audio content.”
Features of Supertone Play
Supertone Play is currently available in English, Korean, and Japanese, with plans to expand to Spanish and Chinese later this year, as stated in a press release on February 18.
This new tool offers a library of 150 voice characters designed for a variety of content creation purposes, including natural conversation, professional narration of news, audiobooks, and “Voices of memes controlled by the trend.”
Additionally, it features a voice cloning function, allowing users to create a voice model from their own speech in just 10 seconds. The voice character can be adjusted to simulate different emotional tones.
For those wary of misuse, Supertone ensures that the technology includes safety measures such as Voice Kapcha, which guarantees that only the user’s voice is utilized for creating voice models. It also embeds illegible watermarks to monitor any vocal synthesis created through the platform.
Quality and Accessibility
The company asserts that the tool produces high-quality voices suitable for video games, commercials, videos, and podcasts. Users can currently access the tool on the Supertone site and can try it for free until April 16.
“Play represents a significant step forward in making AI Voice technology more accessible and universal for creators worldwide,” stated Supertone’s President Kiogo Lee.
“With its highly expressive and natural speech synthesis, the game empowers creators to bring their stories to life in new and dynamic ways. Whether it’s content creation, storytelling, or immersive experiences, we are excited to provide a tool that enhances creative expression through Voice AI.”
“Play notes a big step forward in making AI Voice technology more affordable and universal for creators around the world.”
Kiogo Lee, Supertone
During beta testing, the tool was utilized in over 80 countries, generating around 260,000 audio sentences, with approximately 120,000 or 42% being downloaded, indicating strong user engagement with the generated content.
HYBE acquired Supertone in 2022 to enhance music creation capabilities through AI-generated vocals.
Within less than a year, Hybe-signed artist Midnatt released a track titled Masquerade, recognized as the first track in history to feature AI-generated vocals in multiple languages, including Korean, English, Japanese, Chinese, Spanish, and Vietnamese.
Last year, AI introduced Supertone Shift, a tool that enables artists to modify and clone their vocals in real-time, for instance, to convert an artist’s voice into a chorus during a live performance.
Additionally, Hybe recently debuted a virtual pop group, Syndi8, featuring vocals created by Supertone.
In a demonstration of Supertone’s capabilities during an earnings call last year, Hybe revealed that an investor presentation previously delivered by then-performer Jivon Park was actually executed through the AI Voice tool, which mimicked the lead’s vocals.
Lucas focuses on the interaction between music and society. He specializes in how music influences and reflects cultures, analyzing the role of sound in social, political, and cultural events around the world.
