Amazon Polly adds 14 generative voices and two languages
AWS has updated Amazon Polly’s Generative speech engine with a new model aimed at more natural and accurate output, adding 14 voices and making Hindi and Finnish generally available.
AT A GLANCE
Key points
- The Generative engine now runs an updated speech model.
- AWS added 14 voices across nine locales.
- Hindi and Finnish are now generally available on the Generative engine.
- Low-latency streaming and existing SSML support remain available.
What is changing?
AWS updated Amazon Polly’s Generative engine with a new model on October 20. The company says the upgrade is designed to deliver more natural and accurate speech while retaining low latency, with particular improvements targeted at complex content such as numbers, email addresses and street addresses.
The release adds 14 voices across nine locales. The portfolio covers US and Australian English, Hindi and Indian English, German, US Spanish, Brazilian Portuguese, Korean and Finnish. Hindi and Finnish are generally available on the Generative engine for the first time.
Existing SSML tags remain available for controlling supported speech characteristics. AWS also highlights the low-latency streaming API for contact centers, chatbots and agentic voice applications.
This is not an independent voice-quality benchmark; the feature and quality claims come from AWS’s product announcement. Production teams should validate pronunciation, latency and consistency on their own languages and content.
Explore the original sources
Source date: 20 October 2026.
Published by AI Radar: 23 Oct 2026, 01:09 (Istanbul, UTC+3). Updated: 23 Oct 2026, 14:07.
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…