Sinsy

Sinsy (Singing Voice Synthesis System) (しぃんしぃ) is an online Hidden Markov model (HMM)-based singing voice synthesis system by the Nagoya Institute of Technology that was created under the Modified BSD license.[1]

Sinsy
Developer(s)Nagoya Institute of Technology
Stable release
0.92 / December 25, 2015 (2015-12-25)
Preview release
3.9 / 2017/12/25
Operating systemLinux
Available inJapanese, English, Chinese
TypeVocal Synthesizer Application
LicenseModified BSD license
Websitewww.sinsy.jp

Overview

The online demonstrator is free to use, but will only generate tracks up to 5 minutes. The user uploads data in the MusicXML format, which the Sinsy website reads to output a WAV file of the generated voice. Gender factor, vibrato intensity, and pitch shift can be adjusted prior to output.[2]

As of December 25, 2015 the official creators of the Sinsy were Keiichi Tokuda (Producer and designer), Keiichiro Oura (Design and Development),[3] Kazuhiro Nakamura (Development and Main Maintainer), and Yoshihiko Nankaku.[4]

It was originally only in Japanese and English but Mandarin was later added, the website only supports languages for English and Japanese despite this currently.[5][6]

In 2016, Sinsy stated using "Deep learning" processing known as "DNN".[7]

Products
  • Yoko (謡子), Japanese female vocal, she has two vocals for the service right now both in Japanese, one being a beta and the other a fully released version.
  • Xiang-Ling (香鈴), Japanese female vocal, an English vocal was added Christmas, 2015. Mandarin was added also to her language capabilities.
  • Matsuo-P (松尾P), English masculine vocal.
  • Namine Ritsu S (波音リツＳ), a Japanese male vocal currently in beta. Originally produced for UTAU, it released on December 25, 2013.
  • Unidentified; an unknown vocal in Japanese
  • Unidentified; a second unknown vocal in Japanese
gollark: It's not entirely unreasonable of Apple, though. If they did NOT slow down old phones the batteries would just run out stupidly fast. Not documenting it is bad, though.
gollark: > LineageOS is a pain to installNot really, you just need to flash TWRP or something then flash the LineageOS image using that.
gollark: On Android devices, not iPhones.
gollark: You could run LineageOS or something on it, probably.
gollark: New phones cost more than $300 or so, no? Or at least the phone you got.

References
  1. Ju-Yun Cheng, Yi-Chin Huang, Chung-Hsien Wu (2014-12-04). "HMM-based Mandarin Singing Voice Synthesis Using Tailored Synthesis Units and Question Sets" (PDF). Association for Computational Linguistics and Chinese Language Processing. Retrieved 2017-05-17.CS1 maint: multiple names: authors list (link)
  2. Hentai (2012-12-27). "Sinsy Updates to 3.3 & Releases English Demo". Engloids.Info. Retrieved 2015-06-04.
  3. "Keiichiro Oura". Nagoya Institute of Technology. Retrieved 2017-05-17.
  4. "The HMM-Based Singing Voice Synthesis System "Sinsy" version 0.92". Sinsy.sourceforge.net. Retrieved 2016-01-28.
  5. ITmedia ニュース - 初音ミクとも簡単に対話できる「MMDAgent」、その詳細を聞いてきた. Retrieved November 23, 2013
  6. Nakamura, K.; Oura, K.; Nankaku, Y.; Tokuda, K. (May 2014). "HMM-Based singing voice synthesis and its application to Japanese and English". 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP): 265–269. doi:10.1109/ICASSP.2014.6853599.
  7. https://www.itmedia.co.jp/news/articles/1910/02/news087_2.html
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.