silero-models
snakers4/silero-models
Silero Models: pre-trained text-to-speech models made embarrassingly simple
speech-synthesistext-to-speechpretrained-modelspytorchtorch-hubspeech-to-textrussian-languagemultilingual-ttsmachine-learningspeech-processing
Summary
Silero Models is a popular open-source repository providing pre-trained text-to-speech (TTS) and speech-to-text models for multiple languages, with a strong focus on Russian and other CIS languages. It is designed for simplicity and efficiency, offering one-line inference via PyTorch Hub or a pip package. The project features a variety of high-quality, fast neural TTS models that support SSML, automated stress marking, and homograph disambiguation for Russian.