Deep Voice 3: Ten Million Queries on a Single GPU Server
Although much of the attention around deep learning for voice has focused on speech recognition, developments in artificial speech synthesis (text to speech) based on neural network approaches have been just as swift. …