ToolsCompaniesAbout

Amazon Polly

by Amazon@amazonSeattle, Washington and Arlington, Virginia, USANo ratings yet

Cloud text-to-speech API that converts text into lifelike audio

Visit aws.amazon.com
Tool type
Cloud text-to-speech service accessed through APIs
Made for
Developers building voice applications
Runs on
AWS cloud, accessed through APIs and SDKs
Setup is done in the AWS Console; the API is called from your application code, not used as an end-user web app.
Last checked
28 Oct 2026
3 product sources

What it does

Amazon Polly takes text as input, including articles, web pages, and PDF content, and returns synthesized speech as an audio stream. Output can be stored in standard audio formats such as MP3 and OGG, at sample rates of 8,000, 16,000, or 22,050 Hz. Speech can also be produced as Speech Marks metadata, which describes timing for features like lip synchronization or highlighted text.

People typically use Amazon Polly by calling its API from their own applications, such as adding narration to RSS feeds, websites, videos, or IoT devices, or building voice response systems. Custom lexicons adjust pronunciation of acronyms and product names, and SSML tags control phrasing, emphasis, and intonation. Several voice engines are available, including Standard, Neural, Long-Form, and Generative, each billed differently.

Amazon Polly suits developers and businesses building voice features into applications, though it is a cloud API rather than an end-user app. Using it requires an AWS account and billing setup, and usage is metered per character, with free allowances that vary by engine and are limited in duration. Amazon Polly is not open source, and its code cannot be downloaded or self-hosted.

What it can do

  • Converts text, including articles, web pages, and PDFs, into speech audio
  • Offers dozens of lifelike voices across 40+ languages and variants
  • Customizes pronunciation with custom lexicons for acronyms and company terms
  • Controls phrasing, emphasis, and intonation with SSML tags
  • Outputs MP3, OGG, and other standard audio files, plus Speech Marks metadata
  • Caches and replays generated speech without additional cost

Works with

  • AWS SDKs for Android, JavaScript, iOS, Java, and .NET

Example workflows

Narrating a website article
Input: article text from a web page. Steps: send the text to the Polly API with a chosen Neural voice, optionally apply a custom lexicon for brand names. Output: an MP3 file stored for playback on the page.
Voice response in an IVR system
Input: scripted prompts for callers. Steps: generate speech for each prompt through the API, cache the audio files, and replay them during calls. Output: stored audio clips played by the phone system.

Before you choose

  • Pricing is per character processed, and different voice engines have different per-million-character rates.
  • Free-tier allowances differ by engine and are time-limited (for example, 12 months for Neural voices), so they are not a permanent free tier.
  • Polly is a cloud API; there is no open-source version to self-host, and output audio must be generated through AWS.

Listed under

Get started

Getting started page links to sign-up and to the developer guide and code samples.

  1. Create an AWS account at the sign-up link and sign in to the AWS Console.
  2. Open Amazon Polly in the console and review the Getting Started guide and developer documentation.
  3. Install the AWS SDK for your programming language and call the Polly API with text, a voice, and an output format.
Sign up for a free AWS account

Plans and prices

Prices in USD

Billed monthly for characters processed. Free allowances are time-limited for some engines (Neural 1M characters/month for 12 months; Long-Form 500K for 12 months; Generative 100K for 12 months; Standard 5M/month). New AWS accounts can also receive Free Tier credits from July 15, 2025, usable on Polly within stated time limits. Confirm current rates on the pricing page.

Standard voices$4 per 1M charactersOutside free tier; speech or Speech Marks requests.
  • 5M characters/month free tier
  • Speech or Speech Marks requests
Neural voices$16 per 1M charactersOutside free tier; speech or Speech Marks requests.
  • 1M characters/month free for 12 months
  • Speech or Speech Marks requests
Long-Form and Generative voices$100 per 1M characters, Long-FormOutside free tier; Long-Form speech or Speech Marks requests.$30 per 1M characters, GenerativeOutside free tier; Generative speech requests.
  • Long-Form: 500K characters/month free for 12 months
  • Generative: 100K characters/month free for 12 months

Source: aws.amazon.com/polly/pricing, checked 28 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 28 Oct 2026. Each source supports part of the overview; availability and prices can change.

Ratings and reviews

No ratings for Amazon Polly yet.

Be the first to rate it

What’s changed

Free account

See what’s changed at Amazon Polly

Launches, price moves and pivots, each with a date and a source. Free with an account.

Signals

Free account

See Amazon Polly’s signals

Launch traction, hiring and release activity, tracked over time. Free with an account.

Data and security

Free account

See how Amazon Polly handles your data

Training on your files, compliance, hosting and the models behind it. Free with an account.

Compare with similar tools

Free account

Compare Amazon Polly side by side

Prices, platforms and features against similar tools, with sources. Free with an account.

Where this comes from

How we research
Company website
aws.amazon.com: product, pricing and policy pages
YC directory mirror
Batch and company name
Web search
Confirms the official site
Social accounts
Accounts the company links, and its posts on X
Human review
Checked by a person
knowledge_base

Last checked 28 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.