Versions:
Deepvocal Hitsuboku Kumi Chinese voicebank (dvtb2) version 1.1.0 is a single-version speech synthesis dataset published by oxygendioxide that ports the original Hitsuboku Kumi Chinese voicebank created by cubialpha into the Deepvocal engine. Designed for the Audio Production category, the dataset provides a complete set of Chinese-language phonetic samples and timbral characteristics necessary to reproduce the distinct youthful female voice of the virtual singer Hitsuboku Kumi within Deepvocal-compatible projects. Sound engineers, cover producers, and independent musicians load the voicebank into the Deepvocal editor to generate sung Mandarin lyrics, automate pitch curves, and layer vocal tracks without requiring a live performer. Typical use cases include creating original Chinese pop songs, replacing guide vocals in demo arrangements, producing multilingual crossover covers that blend Kumi’s tone with Japanese or English voicebanks, and rapid prototyping of vocal melodies during composition sessions. Because the port retains the original pronunciation style recorded by cubialpha, users obtain consistent articulation of initials, finals, and tonal transitions that align with contemporary Chinese vocal synthesis standards. The package integrates seamlessly with Deepvocal’s parameter automation system, allowing real-time manipulation of vibrato, breathiness, and consonant velocity to match diverse musical genres from ballads to up-tempo electronic tracks. Released in a single iteration, version 1.1.0 represents the finalized dataset after optimization for the Deepvocal runtime, ensuring stable phoneme mapping and reduced spectral artifacts across the supported pitch range. The software is available for free on get.nero.com, with downloads provided via trusted Windows package sources such as winget, always delivering the latest version and supporting batch installation of multiple applications.
Tags: