app.py

import torch
import gradio as gr

language = 'ru'
model_id = 'v3_1_ru'
sample_rate = 48000
speaker = 'baya'  # Женский голос

# Скачиваем и загружаем модель Silero
model, _ = torch.hub.load(repo_or_dir='snakers4/silero-models',
                          model='silero_tts',
                          language=language,
                          speaker=model_id)

def tts(text, speaker=speaker):
    audio = model.apply_tts(text=text, speaker=speaker, sample_rate=sample_rate)
    return (sample_rate, audio)

iface = gr.Interface(
    fn=tts,
    inputs=gr.Textbox(label="Введите текст"),
    outputs=gr.Audio(label="Голос"),
    title="Русский женский голос — Silero TTS"
)

if __name__ == "__main__":
    iface.launch()