eSpeak NG
| eSpeak NG | |
|---|---|
| Basic data
|
|
| Maintainer | Jonathan Duddington, Reece Dunn |
| developer | 2006 (fork 2010) |
| Current version |
1.50 ( December 2, 2019 ) |
| operating system | Unixoide, Windows |
| programming language | C. |
| category | Speech synthesizer |
| License | GNU GPL v3 + |
| github.com/espeak-ng/espeak-ng | |
eSpeak NG is a compact open source speech synthesizer for Linux, Windows and other platforms. It uses a formant synthesis method that provides many languages in a small size. A large part of the programming for the language support of eSpeak NG takes place via rule files with feedback from native speakers.
Because of its small size and many languages supported, it is included as the default speech synthesizer in the NVDA screen reader for Windows, Android, Ubuntu, and other Linux distributions. Its predecessor eSpeak was used by Google Translate in 2010 for 27 languages; 17 of these were later replaced by commercial voices.
The quality of the language voices varies greatly. In eSpeak NG's predecessor, eSpeak, the first version of each language was based on information from Wikipedia. Some languages have received more work or feedback from native speakers than others. Most of the people who have helped improve the various languages are blind users of text-to-speech software.
history
In 1995, Jonathan Duddington launched the Speak speech synthesizer for RISC OS computers that support British English. On February 17, 2006, Speak 1.05 was released under the GPLv2 license, initially for Linux, with a Windows SAPI 5 version added in January 2007. The development of Speak continued until version 1.14 when it was renamed eSpeak.
The development of eSpeak was continued from version 1.16 (there was no release 1.15 yet) and supplemented by an eSpeakEdit program for editing and creating the eSpeak voice data. These were only available as separate source and binary downloads up to eSpeak 1.24. Version 1.24.02 of eSpeak was the first version of eSpeak to be version-controlled using subversion, with separate source and binary downloads made available on Sourceforge . From eSpeak 1.27, eSpeak has been updated to use the GPLv3 license. The last official eSpeak version was 1.48.04 for Windows and Linux, 1.47.06 for RISC OS and 1.45.04 for Mac OS X.11, the last development version of eSpeak was 1.48.15 on April 16, 2015.
On June 25, 2010, Reece Dunn started an eSpeak fork on GitHub with version 1.43.46 . This started with the intention of making eSpeak easier on Linux and other POSIX platforms. On October 4th, 2015 (6 months after the publication of eSpeak 1.48.15) this spin-off began to differentiate itself more strongly from the original eSpeak.
On December 8, 2015, there were discussions on the eSpeak mailing list about Jonathan Duddington's lack of activity in the past eight months since the last eSpeak development release. This led to discussions about the further development of eSpeak in Jonathan's absence. The result was the development of the espeak-ng (Next Generation) fork, which uses the GitHub version of eSpeak as the basis for future development.
The eSpeakg NG fork was started on December 11, 2015. The first release of eSpeak NG was 1.49.0 on September 10th, 2016, with significant code cleanings, bug fixes and language updates.
Installation programs for Windows can be downloaded from GitHub. New languages can be added manually. Python - wrappers are available.
Individual evidence
- ↑ Release 1.50 . December 2, 2019 (accessed December 3, 2019).
- ↑ Canonical: Ubuntu Manpage: espeak-ng - A multi-lingual software speech synthesizer. Retrieved November 11, 2018 .
- ↑ espeak-ng (1) - espeak-ng - Debian stretch - Debian man pages. Retrieved November 11, 2018 .
- ↑ How to Add a Language to eSpeak NG. Retrieved November 11, 2018 .
- ↑ py-espeak-ng. Retrieved November 11, 2018 .