ddo-tts

Project Overview

ddo-tts is a speech engine with EdgeTTS built in, using rany2/edge-tts to provide Microsoft Edge read-aloud for reading apps such as Legado, with no extra service to deploy. It also integrates Xiaomi MIMO speech synthesis for use cases like voice cloning. Compared with common TTS engines, it changes how audio is buffered, handling it in memory and releasing read media after each chapter.

Install and listen: the engine is built in, with no separate deployment.

Key Features

• Built-in EdgeTTS: Microsoft Edge read-aloud via rany2/edge-tts, ready out of the box
• Xiaomi MIMO synthesis: integrates Xiaomi MIMO speech synthesis for capabilities such as voice cloning
• In-memory buffering: audio streams are buffered in memory instead of constantly writing and deleting disk files
• Per-chapter release: read media is freed after each chapter to lower memory use
• Install and listen: no extra service to set up; use it for reading right after installation

Target Audience

• Users who listen to books in Legado
• Users who want to swap the reading engine in their reading app
• Power users who care about local resource usage

Quick Start

1. Get and install the ddo-tts engine
2. Switch the reading engine to ddo-tts in Legado
3. Choose an EdgeTTS or Xiaomi MIMO voice
4. Open an eBook and start reading aloud to listen

Core Advantages

• No deployment: EdgeTTS is built in, so no self-hosted speech service is needed
• Low footprint: audio is buffered in memory and released per chapter, cutting frequent disk I/O
• Extensible: beyond EdgeTTS, Xiaomi MIMO synthesis is also available

System Requirements

Item	Requirement
Platform	Android, used with reading apps such as Legado
Voice sources	Microsoft EdgeTTS, Xiaomi MIMO
Network	Access to the relevant APIs when using online synthesis

Notes

• Online synthesis depends on third-party APIs, so changes there can affect availability
• Voices and quotas may be limited by the provider's policies
• For personal study and personal use only