Amazon Polly

What Amazon Polly does

• Multiple voice engines. Standard, Neural, Long-Form and Generative engines are offered, each priced separately per character.
• SSML support. SSML tags adjust emphasis, intonation and phrasing of the generated speech.
• Custom lexicons. Lexicons set pronunciation for acronyms, brand names and internal terms.
• MP3 and OGG output. Audio is returned in MP3 and OGG formats, and can be cached and replayed at no extra cost.
• Speech Marks metadata. Speech Marks metadata can be requested alongside audio and is billed per character.

Who it’s for

• Developers adding spoken output to apps via an API
• IVR and voice-response systems
• Media voiceovers generated from text at scale

Amazon Polly pricing

Pay per character: Standard $4, Neural $16, Generative $30, Long-Form $100 per 1M characters. Time-limited free tier for new AWS accounts.

As listed on November 9, 2026. Check aws.amazon.com for current prices.

Things to know

• Not open source; the code cannot be downloaded or deployed in your own environment
• Free tier is limited: most engines include free usage only for the first 12 months
• Delivered as an API; no standalone consumer app is listed on the product page

Quick answers

What is Amazon Polly?

Amazon Polly is a text-to-speech service from Amazon Web Services that converts text into spoken audio for developers building applications.

Is Amazon Polly free?

Amazon Polly is billed by usage. (As listed on November 9, 2026.)

Who makes Amazon Polly?

Amazon Polly is made by Amazon Web Services.

Where can you use Amazon Polly?

Amazon Polly is available on an API for developers.

What are some alternatives to Amazon Polly?

Other tools filed in the same section include Murf AI, WellSaid and NaturalReader.

Tags

#text-to-speech#aws#ssml#neural-voices#api#voiceover