Google said the translation is from the first to last model which can translate directly from one language to another.

Google Unverils Translatotron, Its Speech-to-Speech Translation System.

Google has announced its first direct speech-to-speech translation system called “Translation” that can convert verbal communication from one language to another while maintaining the speaker’s voice and tempo.

The translation is a sequence-by-order network-based source that draws source characters – a visible representation of frequencies – Creates characters as input and target language content, Ye Jia and Ron Weiss, a Google-based software engineer, wrote a blog post intelligence (AI) on Wednesday. Wrote.

The model uses two separate trained components – a neural vocoder that converts output spectrogram into a time-domain waveform and a speaker encoder, which can be used to maintain the character of the voice speaker in a synthesized translation speech.

For now, the results of Google’s translation system are behind a conventional cascade system, but the search engine giant said it “demonstrates the possibility of speech translation from the last direct live conversation”.

Also Read :-Google Play Store Ban China App Developer With 600 Million Downloads From.


“Includes Speaker Encoder Network, Translation is able to maintain the voice speaker features of the speakers in the translated lecture, which makes the translated lecture more natural and less flashy,” said Jia and Weiss.

The company said that the translation is the last of the last model which can translate directly from one language into another language.

“We hope this work can serve as a primary point of view for future research in the last lecture-to-speech translation method,” the blog post said.

Recently, Google is working aggressively in its translation process.

To add more language to its real-time translation feature in 2018, the company introduced accents with different language-based languages ​​in Google’s translation.

In February, Google’s users expanded the “interpreter” mode on the home speakers and other third-party devices so that users can translate real-time conversations with compatible devices and translate them into 26 listed languages.

For the latest Technews and reviews Follow on FacebookInstagramTwitterPinterest