TalkifyTTS

云端大模型驱动的 Android 语音合成应用(TTS引擎)。支持ChatGPT、Gemini、豆包、腾讯、微软、千问等模型。Cloud LLM-driven Android TTS engine supporting ChatGPT, Gem

What is TalkifyTTS?

云端大模型驱动的 Android 语音合成应用(TTS引擎)。支持ChatGPT、Gemini、豆包、腾讯、微软、千问等模型。Cloud LLM-driven Android TTS engine supporting ChatGPT, Gemini, Doubao, Tencent, Microsoft, Qwen, and more.

Overview of TalkifyTTS

What it is

TalkifyTTS is an Android TTS engine that connects to cloud large language models and a local AI speech synthesis engine. It provides high-quality, expressive speech synthesis through the Android standard Text-to-Speech interface. The app supports multiple providers including ChatGPT, Gemini, Doubao, Tencent, Microsoft, Qwen, and more.

Key features

  • Multi-provider support: built-in access to Alibaba Qwen, Volcengine Doubao, Tencent Cloud, Microsoft Azure, MiniMax, Xiaomi MiMo, Google Gemini TTS, and OpenAI, with one-click switching.
  • Local offline synthesis: built-in ZipVoice local AI engine with 12 selected Chinese voices, no API key needed, zero cost, works offline.
  • Custom configuration: supports custom API base URL and model ID for each provider, HTTP/SOCKS proxy for Google and OpenAI suppliers.
  • Style instructions: control emotion, tone, and speed via natural language description (supported by Xiaomi MiMo, Google Gemini TTS, OpenAI).
  • System-level integration: implements standard Android TTS interface, works seamlessly with reading apps like Legado and Google Play Books.
  • Streaming synthesis: uses streaming technology to reduce first-word latency for near-real-time response.
  • Stable background: guidance for network, battery optimization, and notification permissions to ensure uninterrupted long reading sessions.

Use cases

Enhancing text-to-speech for reading applications, such as e-book readers and audiobook apps. The cloud providers offer high-quality voice synthesis for various languages, while the local offline engine provides privacy-friendly, cost-free synthesis for personal non-commercial use. Suitable for users who want expressive, natural-sounding speech from their Android device without mechanical-sounding voices.

Getting started

To build and run from source:git clone https://github.com/LonePheasantWarrior/TalkifyTTS.gitcd TalkifyTTS./gradlew assembleDebugFor detailed setup of each provider (API keys, proxy configuration), see the repository README.

License and pricing

The project code is released under the MIT License. Third-party dependencies and model assets licenses are noted in NOTICE.md. The ZipVoice model weights are trained on the Emilia dataset (CC-BY-NC-4.0), and local synthesis is for non-commercial use only.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/lonepheasantwarriortalkifytts.svg)](https://yunodb.com/tools/lonepheasantwarriortalkifytts?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/lonepheasantwarriortalkifytts?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/lonepheasantwarriortalkifytts.svg" alt="Featured on YunoDB" height="20"></a>