Underdog’s New 32MB AI Voice Model Works Without Internet
AI startup Underdog has released Woof, a new text-to-speech model that is just 32MB in size and can generate speech … Read More The post Underdog’s New 32MB AI Voice Model Works Without Internet appeared first on ProPakistani .
Underdog, an AI startup, has introduced Woof, a compact text-to-speech model with just 32MB in size. This innovative model enables speech generation directly on devices, such as smartphones and computers, without the need for cloud servers. An interactive browser game, titled Escape, showcases Woof's capabilities by converting typed messages into spoken audio.
The game features a retro-themed room with a typewriter, CRT monitor, desk lamp, and music tuner. The introduction to the game highlights Underdog's emphasis on running AI models locally, offering a more private and lightweight alternative to cloud-based voice generation. Previously, Underdog released Bark, a 167MB speech-to-text model, and a 27-billion-parameter AI model designed for laptops.
The company aims to reduce reliance on cloud computing while providing powerful AI tools on existing devices, catering to specific functions like speech generation where large models may not be necessary.
Written by urgent.news from ProPakistani's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.