Amazon Polly launches gated voice cloning from a few minutes of audio
AWS ML Blog · rss · 2026-11-20
AWS announced general availability (gated release) of Amazon Polly voice cloning. With 1-10 recordings from a consenting speaker, CreateVoice builds a custom voice in 5-10 minutes, synthesized by Polly's generative engine with similar naturalness to existing generative voices.
Key points
- Seven languages at launch (English, Spanish, German, French, Italian, Portuguese, Hindi); each cloned voice speaks only the language it was recorded in
- Building a voice is free; usage is pay-as-you-go
- Built-in safeguards: per-voice consent confirmation, watermarked output, account-scoped access via IAM, optional customer-managed KMS encryption; Polly doesn't retain recordings or use them for training, and DeleteVoice removes data immediately
- Use cases: consistent brand voice across IVR/contact centers/ads, updating prompts without studio sessions, scalable audiobooks, courses, podcasts and game NPCs
Available in all regions supported by the Polly generative engine; access is open to select customers, others can apply.
More from Multimodal
- The Tide Machine: a generative art piece concepted by a human and orchestrated by GPT Astra 6.2 — dreamwieber · 2026-11-20
- Minimax H3 Ref2Va + VHS LoRA: recreating 90s videotape aesthetics — Engine_Hungry · 2026-11-20
- Ideogram 4.6 released, and it's the open-weights version — skipping 4.5 — KokaOP · 2026-11-20
- Google opens Project Genie 4 waitlist access to Google AI Pro subscribers — testingcatalog · 2026-11-20
- Tesseract brings 3D spatial audio generation to Claude and ChatGPT agents — Ror_Fly · 2026-11-20
- Open-source Reverb-TTS clones voices 17x realtime, fully local on MacBook M4 Pro via WebGPU — alexcovo_eth · 2026-11-20