Amazon Polly's Generative TTS engine gets a major accuracy and voice upgrade
In short: AWS has upgraded the Amazon Polly Generative engine with more natural, expressive speech and better accuracy for numbers, addresses and confirmation codes. The update adds 14 new voices across 9 locales, bringing the total to 57 generative voices in 11 languages and 25 locales, including first-time general availability for Hindi and Finnish. It is a drop-in upgrade: voice names, APIs and pricing stay the same, so existing applications work unchanged.
This summary was generated automatically by AI from AWS Machine Learning's publication. It is our own text, not a copy of the original — facts, figures and quotes belong to the source, linked above and below.
What changed?
- 114 new generative voices added across 9 locales
- 2Hindi and Finnish generative voices now generally available for the first time
- 3Total catalog now 57 generative voices across 11 languages and 25 locales
- 4Improved accuracy for numbers, currency, addresses and dates
- 5Region expansion to AWS GovCloud (US-West), now 11 AWS Regions total
- 6Existing voice names, SynthesizeSpeech API and pricing unchanged
- 7Bi-Di Streaming API (StartSpeechSynthesisStream) available for real-time voice agents
| Parameter | Before | Now |
|---|---|---|
| Voice catalog | 43 generative voices | 57 generative voices across 11 languages and 25 locales |
| New voices | Not specified | 14 new voices across 9 locales |
| New languages | Not available in Generative engine | Hindi and Finnish generally available |
| API | SynthesizeSpeech API, same voice names | Unchanged; drop-in upgrade, no code changes required |
| Pricing | Standard pay-as-you-go pricing | Same price, no cost increase, no new SKU |
| Region availability | 10 AWS Regions | 11 AWS Regions including GovCloud (US-West) |
| Accuracy | Baseline generative voice accuracy | Improved accurate rendering of numbers, currency, addresses, dates |
| Streaming | Not specified | Bi-Di Streaming API (StartSpeechSynthesisStream) supported for real-time voice agents |
Why it matters
Better entity accuracy (tracking numbers, addresses, confirmation codes) can reduce escalations from IVR to human agents, while more natural prosody improves the conversational feel of automated calls. Because it's a drop-in upgrade with no API or pricing changes, teams get quality gains without migration work.
What it means for AI agents and contact centers
If you run voice agents or IVR flows via Amazon Connect, Genesys, or direct Polly API calls, the upgrade applies automatically with no code changes, so it's worth re-testing call quality, number/address pronunciation accuracy, and latency on the streaming API before and after. The streaming API is particularly relevant for real-time telephony voice agents needing low-latency playback.
🧪 Worth Testing
Free accuracy and naturalness improvements with no integration changes make this low-risk to validate against current TTS choices for IVR and voice agent pipelines, especially where numeric/address readback accuracy affects call resolution rates.
Sources
- AWS Machine LearningOfficialPrimary sourceOriginal article →„Amazon Polly delivers a major upgrade to its Generative engine“22 Oct 2026, 21:06
- Published by source
- 22 Oct 2026, 21:06
- Found by our system
- 22 Oct 2026, 21:30
- Summary generated
- 22 Oct 2026, 21:32
This article was written by AI from the original source. Facts, numbers and prices come from the source; missing values are marked “Not specified”. Legal notice, copyright and privacy