Daniel McKenna - Creating a Text-To-Speech System in Rust
Text-To-Speech (TTS) systems are challenging: touching on topics such as text processing, signal processing, statistics, linguistics, statistics and deep learning. This talk will strip back some of these layers of complexity and discuss the implementation of TTS systems of the past, present and future. Well look at how Emotech uses Rust to implement our next-generation TTS in production and how to implement a TTS system from scratch in Rust (with and without Neural Networks).