Vois is a desktop voice AI studio that generates studio-quality speech entirely on your computer without requiring internet connectivity. The application runs 100% locally, ensuring that nothing leaves your machine while providing professional-grade text-to-speech capabilities.

Vois offers 63 studio-quality voices across 15 character archetypes and supports 23 languages through 3 different TTS engines optimized for fast drafts, expressive English, and multilingual content. The platform includes voice cloning functionality from short audio samples, a script editor with multi-speaker dialogue support, and a multi-track timeline for mixing and arranging audio content. Professional mastering features include LUFS normalization, de-esser, EQ, and limiter tools with export capabilities to WAV, MP3, FLAC, and AAC formats.

The application operates entirely locally using a native Rust backend without Python or Docker dependencies. It features smart caching technology where editing one sentence only requires regenerating that specific chunk, making iterations instant. The fast engine generates audio at 6x real-time speed on Apple Silicon hardware.

Vois eliminates per-character costs and usage limits associated with cloud services while providing instant iteration capabilities through its caching system. Use cases include converting articles, reports, academic papers, and white papers to audio for accessibility, creating custom bedtime stories, converting ebooks to audio for commutes, podcast creation, and game dialogue generation.

The product serves users with dyslexia, ADHD, and other reading challenges who need text-to-audio conversion, content creators wanting to produce podcasts or audio content, and individuals seeking privacy-focused voice generation solutions. The application runs natively on desktop platforms with a Rust-based architecture optimized for performance.