Documentation - Setup & Configuration

Step-by-step guides to set up OpenTypeless, configure STT and LLM providers, customize hotkeys, and get the most out of your voice input workflow.

The OpenTypeless documentation covers everything you need to get started with AI-powered voice input on your desktop. Begin with the Getting Started guide to install the app on Windows, macOS, or Linux and configure your first speech-to-text provider. Then explore the STT Providers section to learn about each supported provider — Deepgram, OpenAI Whisper, Groq Whisper, GLM-4-Voice, AssemblyAI, and SiliconFlow — including setup instructions, pricing, and performance characteristics. The LLM Providers section explains how to connect AI models for text polishing, covering OpenAI GPT, DeepSeek, Google Gemini, Anthropic Claude, and local options like Ollama and LM Studio. Advanced topics include custom dictionary configuration, hotkey customization, polish prompt templates, and building OpenTypeless from source.

Fast Answers

Short routing answers for setup, provider, quota, and troubleshooting intent.

Which docs should I read first?

Start with Getting Started, then the Provider Index, then Ask Anything setup if you want one-off voice questions.

Where do I fix Ask Anything timeout or no popup?

Use the Ask Anything troubleshooting guide, then check Windows microphone permission or macOS Accessibility permission depending on your platform.

Where do I understand cloud words?

Use Ask Anything credit usage and the pricing page. Hosted STT and hosted LLM usage can consume cloud words, while BYOK/local provider paths depend on your own provider account.

Where do I configure BYOK providers?

Use the Provider Index first, then the specific STT or LLM provider page for API keys, base URLs, model names, and setup notes.