Amazon’s Polly Launches Newscaster Voice and Neural Text-to-Speech Feature


Amazon Polly, a cloud-based Text-to-Speech (TTS) service that launched in 2016, has just introduced two new features: Newscaster and Neural Text-To-Speech (NTTS). Polly turns text into human-like speech, allowing users to build speech-enabled products and develop applications. According to Amazon, the technology behind Polly mimics the human brain by leveraging machine learning. From their AWS blog, “Similar to the way human children learn to speak, these systems generate sounds, then improve their speech by listening to recorded natural speech and copying it.”

  • Amazon Polly Newscaster: Amazon Polly imitates the newsperson style, so news media and other content creators can deliver breaking news and original reporting more quickly and accurately.
  • Neural Text-to-Speech (NTTS): Enables Amazon Polly to quickly learn differences in speaking styles to imitate them. As of yesterday, there are 11 voices available: 3 UK English and 8 US English

The service currently supports 29 languages and provides a variety of voices in multiple languages. For instance, the English US version of Polly has 8 different voices as well as 2 Newscaster options, both in real-time and batch mode. There are 5 female voices and 3 males: Ivy, Joanna, Kendra, Kimberly, Salli, Joey, Justin and Matthew.

Amazon Calls Newsperson Style Unprecedented

The Globe and Mail, one of Canada’s top newspapers, is among the first media outlets to use Polly’s Newscaster. Greg Doufas, Chief Technical and Digital Offer at The Globe and Mail, explains in a video below how Polly helps them increase user engagement. “Amazon Polly Newscaster enables us to provide our readers with more features to further their experience with our newspaper. This text-to-voice feature from AWS is miles ahead of anything we’ve heard to date.”

Other early media adopters are Gannett, owner of USA Today, and Trinity Audio, a company enables companies to transform written website content into listenable audio. “Services like Amazon Polly and features like its Newscaster voice help us deliver breaking news and original reporting with increased speed and fidelity worthy of our brands,” Gannett’s Scott Stein, Vice President of Content Ventures told AWS.

Neural Text-To-Speech Development Allows Alexa to Sound More Realistic

Google Launches New Text-to-Speech Cloud Service