sprocket

Voice conversion software - Voice conversion (VC) is a technique to convert a speaker identity of a source speaker into that of a target speaker. This software enables the users to develop a traditional VC system based on a Gaussian mixture model (GMM) and a vocoder-free VC system based on a differential GMM (DIFFGMM) using a parallel dataset of the source and target speakers.

Paper and slide

• K. Kobayashi, T. Toda, "sprocket: Open-Source Voice Conversion Software," Proc. Odyssey, pp. 203-210, June 2018. [paper]

• T. Toda, "Hands on Voice Conversion," Speech Processing Courses in Crete (SPCC), July 2018. [slide]

Conversion samples

• Voice Conversion Challenge 2018
  • HUB Task
  • SPOKE Task

Purpose

Reproduce the typical VC systems

This software was developed to make it possible for the users to easily build the VC systems by only preparing a parallel dataset of the desired source and target speakers and executing example scripts. The following VC methods were implemented as the typical VC methods.

Traditional VC method based on GMM

• T. Toda, A.W. Black, K. Tokuda, "Voice conversion based on maximum likelihood estimation of spectral parameter trajectory," IEEE Transactions on Audio, Speech and Language Processing, Vol. 15, No. 8, pp. 2222-2235, Nov. 2007.

Vocoder-free VC method based on DIFFGMM

• K. Kobayashi, T. Toda, S. Nakamura, "F0 transformation techniques for statistical voice conversion with direct waveform modification with spectral differential," Proc. IEEE SLT, pp. 693-700, Dec. 2016.

Supply Python3 VC library

To make it possible to easily develop VC-based applications using Python (Python3), the VC library is also supplied, including several interfaces, such as acoustic feature analysis/synthesis, acoustic feature modeling, acoustic feature conversion, and waveform modification. For the details of the VC library, please see sprocket documents in (coming soon).

Installation & Run

Please use Python3.

Current stable version

Ver. 0.18.4

Install sprocket

pip install numpy==1.15.4 cython  # for dependency
pip install sprocket-vc

Run example

See VC example

REPORTING BUGS

For any questions or issues please visit:

https://github.com/k2kobayashi/sprocket/issues

COPYRIGHT

Copyright (c) 2020 Kazuhiro KOBAYASHI

Released under the MIT license

https://opensource.org/licenses/mit-license.php

ACKNOWLEDGEMENTS

Thank you @r9y9 and @tats-u for lots of contributions and encouragement helps before release.

Who we are

• Kazuhiro Kobayashi @k2kobayashi [maintainer, design and development]

• Tomoki Toda [advisor]

Changelog

0.18.4 (2020/12/02)

• Support new api for dependency library

0.18.3 (2019/04/24)

• Implement several functions for GMM-based VC and minor bugfix
• #133
• #132
• #130
• #127

0.18 (2017/10/01)

• Release first ver.
• Baseline system for Voice Conversion Challenge 2018