An open-source, on-device voice assistant for Android with speech and graphical replies and a set of built-in skills.
AndraxDev/speak-gpt
An Android voice assistant app that lets you chat with OpenAI and other LLMs using speech input, image recognition and image generation.
Open-source alternative to:ChatGPT
- Stars
- 446
- Forks
- 81
- Open issues
- 10
- Last commit
- 8 days ago
What is it?
SpeakGPT is an open-source Android app that works as a voice-driven client for large language models. Its main supported backend is OpenAI, with partial support for GROQ and Azure, and text generation through OpenRouter, which gives access to models such as Gemini, Claude, Llama, Gemma and Mistral. Users can add a custom API provider and switch endpoints from the settings.
The app covers the usual chat workflow: conversations are stored locally and can be imported or exported, and you can set system messages, activation prompts and model parameters like temperature and topP. Voice input works through Whisper or Google, and image recognition and DALL-E style image generation are included. It also offers a prompts library, a playground, and entry points in the Android context menu and share sheet.
SpeakGPT requires you to bring your own API key, which is stored on the device. The project is distributed on Google Play and its source is available for compilation. The code is written in Kotlin.
Key features
- 01Voice input through Whisper or Google speech recognition
- 02Image recognition and image generation
- 03Works with OpenAI, GROQ, Azure and OpenRouter, plus custom API providers
- 04Local chat history with import and export
- 05Adjustable model parameters such as temperature and topP
- 06Entry points in the context menu and Android share sheet
Use cases
- Talking to an LLM hands-free while on the go on an Android phone
- Asking questions about photos or screenshots using a vision-capable model
- Using your own OpenAI or OpenRouter key as a mobile AI client
- Sending selected text or shared content to a chatbot from other apps
Quick start
- 1
Install SpeakGPT from Google Play
- 2
Open the app and add your API key in Settings (OpenAI or another supported provider)
Things to know
- Android only; there is no desktop or iOS version.
- Requires your own API key from a supported provider, and usage is billed by that provider.
- Some providers have only partial support, and Gemini is reached through OpenRouter rather than directly.
- Support for Android 9, 10 and 11 is being dropped.
In collections
Topics
FAQ
- What is speak-gpt?
- An Android voice assistant app that lets you chat with OpenAI and other LLMs using speech input, image recognition and image generation. SpeakGPT is an open-source Android app that works as a voice-driven client for large language models. Its main supported backend is OpenAI, with partial support for GROQ and Azure, and text generation through OpenRouter, which gives access to models such as Gemini, Claude, Llama, Gemma and Mistral. Users can add a custom API provider and switch endpoints from the settings.
- Is speak-gpt free to use? What license does it have?
- speak-gpt is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
- Is speak-gpt still maintained?
- Actively maintained. The last commit to speak-gpt was on Oct 30, 2026.
- What can I use speak-gpt for?
- Talking to an LLM hands-free while on the go on an Android phone Asking questions about photos or screenshots using a vision-capable model Using your own OpenAI or OpenRouter key as a mobile AI client Sending selected text or shared content to a chatbot from other apps
- How do I install speak-gpt?
- Install SpeakGPT from Google Play Open the app and add your API key in Settings (OpenAI or another supported provider)
- Where can I get speak-gpt?
- The source code and installation instructions are on GitHub: https://github.com/AndraxDev/speak-gpt
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
An Android app to choose, launch, and set a default among installed AI digital assistants like ChatGPT, Claude, and Gemini.
Desktop client for ChatGPT, Claude and other LLMs, running on Windows, macOS and Linux with local data storage.
Android chat app that sends one prompt to multiple LLM providers at once, using your own API keys.
Self-hostable AI second brain that answers questions from the web and your own documents, with custom agents and scheduled automations.
Android app for chatting with GGUF models running locally via llama.cpp, or with remote providers like OpenAI, Anthropic, Ollama and DeepSeek.