Arabic Added To Amazon Polly Text-To-Speech Cloud Service
The move allows developers to create applications that speak in Arabic and build speech-enabled products and services, including cars, internet of things devices, appliances, automated contact centers, language learning platforms, translation apps and newsreaders.
Cloud provider Amazon Web Services yesterday said Arabic has been added to the list of languages spoken by Amazon Polly, the cloud service that converts text into life-like speech.
The move allows developers to create applications that speak in Arabic and build speech-enabled products and services, including cars, internet of things devices, appliances, automated contact centers, language learning platforms, translation apps and newsreaders.
The artificial intelligence-powered Polly uses deep-learning technologies to synthesize speech that sounds like a human voice. Polly’s Arabic-speaking voice is a female voice called “Zeina” that speaks in Modern Standard Arabic (MSA), a standardized version of the language that is more formal than everyday colloquial Arabic, which includes 30 dialects.
“Used by over 400 million people, Arabic is one of the most spoken languages in the world,” AWS said in a post announcing Zeina, which was unveiled at the AWS Summit in Dubai this week. “Zeina follows the MSA pronunciation, which is the common broadcasting standard across the region. MSA might sometimes sound formal because it differs from day-to-day speaking style. However, it’s the linguistic thread that links the Arabic native-speakers worldwide.”
In 2017, AWS announced it planned to open its first data center in the Middle East -- a location in Bahrain with three availability zones – by early this year, so AWS customers and partners could run their workloads and store their data in the region. That data center still is “coming soon.” AWS opened its second Middle East edge network location -- in Fujairah, United Arab Emirates -- last fall, after opening AWS offices in Dubai, UAE, and Manama, Bahrain, in January 2017.
AWS customers who plans to use Zeina’s Arabic voice include the Dubai-based Emirates NBD, one of the largest banking groups in the Middle East, for an AI-powered virtual assistant that provides new voice banking solutions for customers.
“Emirates NBD has been an early mover in the region in introducing an AI-powered virtual assistant, helping customers calling the bank to converse in natural language and access required services quickly,” said Suvo Sarkar, the company’s senior executive vice president and group head of retail banking and wealth management. “We are now integrating Amazon Polly in English with our automated call center for its quality and lifelike voice and to further enhance customer interactions, and looking to integrate Amazon Polly in Arabic soon. Such technologies will also help us improve our internal efficiencies while delivering better customer experiences.”
Pittsburgh’s Duolingo, a free language-learning platform that includes a website and mobile application, is expected to release its previously announced Arabic course for English speakers on May 1.
“Zeina delivers accurate and natural-sounding speech that is important for teaching a language and matches the quality that we’ve become accustomed to using Amazon Polly for the other languages that we offer,” said Hope Wilson, Duolingo’s learning scientist.
Austria’s iTranslate, developer of a translation and dictionary app that offers text translation and enables voice-to-voice conversations in 100-plus languages, also plans to use Zeina, according to AWS.
Amazon Polly now includes 59 voices in 29 languages, from Mandarin Chinese to Welsh, with male and female voices for most, as well as a bilingual Hindi and Indian English voice.
AWS says all text submissions to Polly are protected by secure sockets technology while in transit and stored using RSA encryption. It maintains Polly service logs and text separately, so that the content can't be linked with a customer identity, according to the company, which says it retains Polly data for 14 days before automatically deleting it.