Amazon Polly adds native voice cloning capabilities
In short: Amazon Polly has introduced a new voice cloning feature that builds synthetic voices from as little as two to ten minutes of speaker audio. This tool is now available through a gated release for seven languages and integrates directly with existing AWS cloud workflows for enhanced security and brand consistency.
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?
- 1New API functionality for registering custom voices using 1-10 short audio clips.
- 2Asynchronous processing pipeline that typically completes within 5-10 minutes.
- 3Automatic encryption for all cloned voice assets using AWS KMS.
- 4Mandatory audio watermarking for all output generated via the new voice models.
| Parameter | Before | Now |
|---|---|---|
| Languages supported | Not specified | English, Spanish, German, French, Italian, Portuguese, Hindi |
| Voice creation time | Not specified | 5-10 minutes |
| Required audio duration | Not specified | 2-10 minutes (up to 15 minutes total) |
Why it matters
This feature allows organizations to maintain a uniform brand persona across various touchpoints, such as phone support, mobile apps, and digital assistants. By avoiding studio sessions, developers can quickly update audio prompts or deploy new localized content while ensuring strict control over who can access and use their custom voices.
What it means for AI agents and contact centers
When evaluating this against existing TTS pipelines, focus on the 'generative' engine performance and integration ease. Test the latency impact when replacing standard voices with these custom clones in real-time agent responses. Ensure your workflows account for the asynchronous 5-10 minute setup period when onboarding new voices.
🧪 Worth Testing
Testing is recommended to determine if these custom voices improve user engagement in voice-based agent interactions compared to standard stock voices.
Sources
- AWS Machine LearningOfficialPrimary sourceOriginal article →„Amazon Polly launches voice cloning: A unique voice for your brand“19 Nov 2026, 21:53
- Published by source
- 19 Nov 2026, 21:53
- Found by our system
- 19 Nov 2026, 22:05
- Summary generated
- 19 Nov 2026, 23:05
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