Google voice AI is similar to humans

Google’s voice-generating AI is now indistinguishable from humans

Google voice AI is similar to humans

Google voice AI is similar to humans

Humans have officially given their voice to machines.

A research paper published by Google this month—which has not been peer reviewed—details a text-to-speech system called Tacotron 2, which claims near-human accuracy at imitating audio of a person speaking from text.

The system is Google’s second official generation of the technology, which consists of two deep neural networks. The initial system translates the text in to a spectrogram (pdf), a visual solution to represent audio frequencies around time. That spectrogram is then given into WaveNet, something from Alphabet's AI research laboratory DeepMind, which reads the chart and generates the equivalent audio things accordingly.

The Google researchers also demonstrate that Tacotron 2 can handle hard-to-pronounce words and names, as well as alter the way it enunciates based on punctuation. For instance, capitalized words are stressed, as someone would do when indicating that specific word is an important part of a sentence.

Unlike some core AI research the company does, this technology is immediately useful to Google. WaveNet, first announced in 2016, is now used to generate the voice in Google Assistant. After readied for generation, Tacotron 2 might be a much more strong supplement to the service. However, the system is just qualified to copy usually the one girl voice; to talk like a male or various girl, Google would have to prepare the system again.