#DreamTUI

A terminal-based generative dream engine.

GitHub Repository | Releases

#Overview

DreamTUI transforms your written dreams into full-screen, animated ASCII art experiences. It uses a sophisticated two-stage AI pipeline to interpret the semantics of your dream and generate a procedural visualization that plays out directly in your terminal.

This is an experimental hobby project created to test the concept of AI-driven ASCII visualization. It is currently in a prototype stage (and frankly, it will probably stay that way forever). However, the project is open source and contributions are highly welcome!

#Features

#AI-Powered Interpretation

• Dream Booster: Enhances simple inputs into vivid, descriptive narratives.
• Semantic Analysis: Extracts mood, symbols, tempo, and atmosphere from your text.

#ASCII Rendering Engine

• Layered Composition: Combines background noise, symbols, and distortion effects.
• Procedural Animation: Real-time animation loop reacting to dream parameters.
• Dynamic Distortions: Visual wobbles and shifts representing dream instability.

#Modern TUI (Terminal User Interface)

• Interactive Screens: Dedicated screens for Input, Animation, and Settings.
• Responsive Design: Adapts gracefully to terminal resizing.
• Keyboard Controls: Full keyboard navigation support.

#Flexible AI Providers

Supports multiple LLM providers, configurable via the built-in Settings UI:

• OpenAI
• Anthropic
• Google Gemini
• Groq
• Ollama (Local)

#Getting Started

#Prerequisites

• Bun (v1.0 or later recommended)
• A terminal with TrueColor support.

#Installation

git clone https://github.com/sametcn99/dreamtui.git
cd dreamtui
bun install
bun start

#Usage

#1. The Input Screen

When the app launches, you will see the Input Screen.

• Type your dream: Enter a description of a dream (e.g., "I was flying over a neon city made of glass").
• Submit: Press Enter to start the dream generation process.
• Settings: Press Ctrl+S to open the configuration menu.

#2. The Dream Experience

Once generated, the visualization will start automatically.

• Movement: The ASCII art flows according to the dream's tempo (calm, chaotic, fast, slow).
• Controls:
  • Space: Pause/Resume animation.
  • R: Restart the dream loop.
  • Esc: Exit back to the Input Screen.

#3. Settings Configuration

Access settings by pressing Ctrl+S on the Input Screen.

• Provider: Select your preferred AI backend (e.g., OpenAI, Anthropic).
• API Key: Enter your API key securely (stored locally).
• Model: (Optional) Override the default model (e.g., gpt-4o-mini, claude-3-opus).
• Navigation: Use Tab to cycle fields, Ctrl+S to Save, Esc to go back.

#Technology Stack

• TypeScript
• Bun
• OpenTUI (terminal UI framework)
• Multiple LLM Providers (OpenAI, Anthropic, Google Gemini, Groq, Ollama)