silero-models
snakers4/silero-models
Silero Models: pre-trained text-to-speech models made embarrassingly simple
speech-synthesistext-to-speechpytorchpretrained-modelstorch-hubrussianmultilingualspeechttstts-models
Summary
Silero Models is a popular open-source repository providing pre-trained text-to-speech models for multiple languages, with a strong focus on Russian and other Cyrillic-script languages. It offers an embarrassingly simple interface via PyTorch Hub or a pip package, featuring fast CPU/GPU inference, support for SSML, and automated handling of linguistic complexities like stress and homographs for Russian. The project is well-documented with extensive examples and Colab notebooks.