2023-08-24
GamaTTS
GamaTTS is an experimental articulatory speech synthesizer, derived from Gnuspeech.
Status: (forever) alpha.
License: GNU GPLv3-or-later.
1. Documentation
A concise documentation in HTML is available.
2. Synthesis examples
- The Chaos by Gerard Nolst Trenité (short version) synthesized by GamaTTS 0.9 (using vocal tract model 5).
| English - Male | MP3 |
| English - Female | MP3 |
| English - Large child | MP3 |
| English - Small child | MP3 |
| English - Baby | MP3 |
3. Performance of GamaTTS 0.9 using the english male voice
CPU: Ryzen 5 1600AF (the software uses only one thread).
| Vocal tract model | Speech duration | Synthesis time | Real-time factor |
|---|---|---|---|
| 0 (double prec.) | 420 s | 3.2 s | 130x |
| 1 (single prec.) | 420 s | 2.7 s | 150x |
| 5 (double prec.) | 422 s | 6.0 s | 70x |
Notes:
- The synthesis time includes the initialization.
- Vocal tract model 0 is equivalent to the one in Gnuspeech.
- Vocal tract model 1 is equivalent to model 0 but uses single precision.
- Vocal tract model 5 has more sections and higher internal sampling rate.
4. Components
The multi-platform software is written in C++ and is developed and tested on Linux+GNU x86_64.
- GamaTTS - Command-line speech synthesizer and library.
- GamaTTS:Editor - Articulatory database editor, approximate clone of Monet from Gnuspeech. Uses Qt 5 or 6.
- GamaTTS module for Speech Dispatcher - Speech Dispatcher provides a common interface for speech synthesizers.
[ git repository ]