Microsoft Translator Text API is a cloud-based API that is conveniently integrated into multiple products, tools, and solutions. Discover below how you can use, integrate, and customize this service in your own application, or simply use it within existing Microsoft applications and services. The Translator API can be used on its own or its results can be customized pre- and post-publishing with the Translator Hub and Collaborative Translation Framework (CTF).
Microsoft Translator Speech API uses Deep Neural Network (DNN) system, which improves translation results. It can handle large amounts of inputs and is trained on analyzing hours of incoming audio speech. DNN is a multi-layered, non-linear, latticed learning system. Each output informs the next layer. Its network of neurons (similar our brain), has predefined probability and size dimensions training, but can also change due to information being transferred from another neuron in a connected, hidden layer, before the final activation of the output neuron determines the best word or character option.