VOCALOID6

This is the central, AI-based technology that generates a highly expressive and natural-sounding singing voice. It uses deep learning to analyze and reproduce singing characteristics from real vocalists, resulting in high-quality vocals with smoother pronunciation and improved pitch/tone changes. The engine allows users to transform their computer into a fabulous vocalist with only melody and lyrics input.

This unique feature proposes a new way to create vocal tracks. Users can import audio data of their own singing (in WAV format) and use it to recreate their personal singing style or phrasing using a VOCALOID6-exclusive voicebank. This functionality allows creators to achieve specific expressive nuances without needing to input musical notation, offering superb fidelity.

VOCALOID6 features voicebanks that offer true multilingual functionality. Creators can sing lyrics that contain a mixture of Japanese, English, and Chinese with a single voicebank. This new feature automatically detects the language and applies natural, native-like pronunciation, which grants creators the freedom to write lyrics that transcend traditional language barriers in music production.

New editing tools have been implemented to allow creators to quickly express their ideas and direct the vocalist. Users can freely manipulate elements like vocal accents, vibrato, and the rhythmic feel of the performance by visualizing notes and pitch using curved lines. This gives users precise control over the vocal style, enabling them to customize a unique way of singing.

VOCALOID6 includes new features that streamline the process of producing complex vocal tracks. The vocal work functionality allows users to instantly create essential production techniques such as doubling (layering multiple vocal tracks) and harmony parts. This one-click capability makes it easy to assemble rich, multi-layered vocal parts just as the creator imagined them.