Freedom251 › Text-to-Speech Software
eSpeak NG
by eSpeak NG

What eSpeak NG is
eSpeak NG is a compact, open-source text-to-speech synthesizer for converting text into spoken audio. It supports more than 100 languages and accents, with a development version listing 127; distributed voices can be checked with the espeak-ng --voices command. Formant synthesis keeps language data small and supports high speaking speeds, but speech is less natural or smooth than systems based on human recordings. Multiple adjustable voices are included. eSpeak NG produces WAV output and supports HTML plus partial SSML. It is available as a command-line program, shared library, Windows SAPI5 version, API, and self-hosted software for Android, Linux, macOS, and Windows. MBROLA can serve as a backend using phonemes with pitch and length information. Its C API and ABI are compatible with eSpeak, although some files are not. The project uses the GNU GPL version 3 or later and permits commercial use.
Who it suits
eSpeak NG suits developers and accessibility tools needing compact, multilingual speech synthesis, command-line or library integration, and commercial use. It may not suit projects requiring highly natural recorded-speech quality.
What it does well
- Supports over 100 languages and accents
- WAV output
- Command-line and shared-library interfaces
- Commercial use permitted
- Adjustable voices
What to know first
- Speech is less natural than recorded voices
- SSML support is incomplete
- Some eSpeak files are incompatible
- CMake required for current compilation
Plans
| Plan | Billed and limits | Price |
|---|---|---|
| eSpeak NG | Open-source engine · desktop and mobile ports · GPLv3+ engine license | Free |
As github.com lists them, read 15 Oct 2026. The maker's own page has the final word.
Verdict
eSpeak NG combines broad language coverage, small data, adjustable voices, and several integration interfaces in a free GPL project. Consider its less smooth formant speech and partial SSML support when evaluating output quality.
Facts from eSpeak NG's own pages
- Android app rating
- The Google Play listing shows 3.4 out of 5 stars from 1.67K reviews. · play.google.com, 15 Oct 2026
- API access
- Yes · github.com, 20 Sept 2026
- API compatibility
- The C API remains API and ABI compatible with eSpeak 1.48.15. · github.com, 15 Oct 2026
- Apple App Store rating
- The Apple App Store listing shows 4.0 out of 5 from 21 ratings. · apps.apple.com, 15 Oct 2026
- Audio and markup
- It can produce WAV output and supports SSML (not completely) and HTML. · github.com, 12 Oct 2026
- Audio output
- The software can generate speech as WAV audio files. · github.com, 15 Oct 2026
- Build requirement
- The 1.52 release notes say it is the last release supporting the Autoconf build system and instruct users to use CMake to compile eSpeak NG. · github.com, 12 Oct 2026
- Build requirements
- The current native build requires CMake 3.10 or later and a C99-capable compiler such as GCC or Clang. · github.com, 15 Oct 2026
- Command line
- Command-line programs speak text supplied as files or standard input on Linux and Windows. · github.com, 15 Oct 2026
- Commercial use
- Yes · github.com, 20 Sept 2026
- Compact footprint
- The executable, language data and many voices total only a few megabytes. · github.com, 15 Oct 2026
- Compatibility
- Its C API and ABI are compatible with eSpeak, while some voice, dictionary, and phoneme files are incompatible with eSpeak. · github.com, 12 Oct 2026
- Component licenses
- Fedora identifies GPL-3.0-only, GPL-3.0-or-later, Apache-2.0, BSD-2-Clause, Unicode-DFS-2016 and CC-BY-SA-3.0 components in the package. · packages.fedoraproject.org, 15 Oct 2026
- Debian interface
- The Debian man page documents speaking text, reading files or standard input, phoneme output and voice listing options. · manpages.debian.org, 15 Oct 2026
- Developer documentation
- Documentation includes user, building, contributor, language and voice-development guides plus C-library guidance. · github.com, 15 Oct 2026
- Export formats
- WAV · github.com, 20 Sept 2026
- Founded
- 2015 · github.com, 28 Sept 2026
- Free plan
- Yes · github.com, 20 Sept 2026
- Independent Android analysis
- AppRecs rates the Android listing 2.9 and reports a trustworthiness score of 86 out of 100 from its analyzed review sample. · apprecs.com, 15 Oct 2026
- Integration
- It is available as a command-line program, a shared library, and a Windows SAPI5 version for compatible screen readers and other programs. · github.com, 12 Oct 2026
- Language coverage
- The development version lists 127 languages and accents; the project says to run `espeak-ng --voices` to check the distributed version. · github.com, 12 Oct 2026
- Latest release
- GitHub lists version 1.52.0 as the latest release, dated December 12, 2024. · github.com, 15 Oct 2026
- Library API
- A shared library and Windows DLL expose a C API for embedding in other programs. · github.com, 15 Oct 2026
- License
- The main eSpeak NG engine is released under the GNU GPL version 3 or later. · github.com, 15 Oct 2026
Where eSpeak NG stands on Freedom251
- Text-to-Speech Software#3 of 223
- Text-to-speech tools#5 of 232
