Skip to content

Latest commit

 

History

History
29 lines (17 loc) · 1.63 KB

README.md

File metadata and controls

29 lines (17 loc) · 1.63 KB

polish-nnsvs-voicebank-albert

This is an NNSVS library for the Polish language created using a dataset recorded from a classically trained baritone. It's intended to use with a compatible software like ENUNU or OpenUTAU. References on how to use the voicebank can be found in the readme.txt and SzopaTatyJarka PL NN Phoneme set.txt. The library also comes with 5 example score files to test the voicebank on.

Example render

laleczka_z_saskiej_porcelany__20220422203657.mp4

Usage example in UTAU scr1

Usage example in OpenUTAU scr2

It's recommended to use the voicebank with OpenUTAU as it allows for flexible vocal editing, live playback and more. If using UTAU, ENUNU ver.0.4.0 is recommended.

This library also supports Japanese hiragana input and allows for Japanese singing with an accent.

Short information about the voicebank

  • Language: Polish
  • Recorded pitch range: A2 ~ E4
  • Recorded tempo range: 78 ~ 130 BPM
  • Data amount: 1:17:13 / 22 songs

More statistics can be found in the STATS.ODS included with the voicebank

Additionally, we included a fan-art package, which contains voice character interpretations from various artists. Those artworks are included for fun and to commemorate the drawings made by the community.