---
title: TTS Server
emoji: 🏃
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 4.38.1
app_file: app.py
pinned: false
---

# TTS Server

The TTS (Text-to-Speech) Server is a Flask-based application that provides an API for synthesizing speech from text using various models. It supports multiple TTS models fetched from Hugging Face Model Hub.

## Features

- **Model Management**: Fetches available TTS models dynamically from Hugging Face Model Hub.
- **Speech Synthesis**: Converts text input into speech audio using selected models.
- **RESTful API**: Provides endpoints for fetching available models and synthesizing speech.
- **Error Handling**: Logs detailed error messages and returns appropriate HTTP status codes for easy debugging.

### Endpoints

- **`/models`**: GET request to fetch available TTS models.

  Example response:
  ```json
  [
      {"repo_id": "csukuangfj/vits-piper-en_US-lessac-medium", "filename": "en_US-lessac-medium.onnx"},
      {"repo_id": "csukuangfj/vits-piper-en_US-hfc_female-medium", "filename": "en_US-hfc_female-medium.onnx"},
      {"repo_id": "csukuangfj/vits-piper-en_GB-southern_english_female-medium", "filename": "en_GB-southern_english_female-medium.onnx"}
  ]