Google says its Parallel Tacotron model generates synthetic voices 13 times faster than its predecessor

In December 2016, Google released Tacotron 2, a machine learning text-to-speech (TTS) system that generates natural-sounding speech from raw transcripts. In a new paper, researchers at the search giant claim to have addressed this limitation with what they call Parallel Tacotron, a model that’s highly parallelized during training and inference to enable efficient voice generation on less-powerful hardware.

Read More