Amazon Polly features

Amazon Polly makes it easy to request an additional stream of metadata that provides information about when particular sentences, words and sounds are being pronounced. Using this metadata stream alongside the synthesized speech audio stream, you can now build your applications with an enhanced visual experience, such as speech-synchronized facial animation or karaoke-style word highlighting.

Please visit the documentation to learn more about how to use Speech Marks. 

Amazon Polly can be used to synthesize speech as if it is were spoken by a TV or Radio newscaster. This can be a great way to read news articles or deliver flash briefing updates. The Newscaster style is currently available for the US English (en-US) Matthew and Joanna voices, British English (en-GB) Amy and US Spanish (es-US) Lupe voice using Neural text-to-speech. Listen to an audio sample in US English, British English or US Spanish.

Amazon Polly enables you to automatically adjust the speech rate based on a maximum allotted amount of time you define with a feature called time-driven prosody. This is beneficial for many use cases, especially when it comes to localization.

For example, suppose you have US English speech embedded in your training video and want to localize this video into German. Let’s say you translate the text using Amazon Translate and voice it with Polly. It is essential that the localized German speech streams in corresponding frames of the video, so the German speech cannot be longer than the US English speech. You can use this feature to more easily facilitate the dubbing process.

Amazon Polly supports all the programming languages included in the AWS SDK (Java, Node.js, .NET, PHP, Python, Ruby, Go, and C++) and AWS Mobile SDK (iOS/Android). Polly also supports an HTTP API so you can implement your own access layer.

Amazon Polly can be accessed via the Polly API (and various language-specific SDKs), AWS Management Console, and the AWS command-line interface (CLI). You have full control over all the capabilities of Amazon Polly, whether you use the service through the console, the API, or the CLI.

With Amazon Polly’s custom lexicons, or vocabularies, you can modify the pronunciation of particular words, such as company names, acronyms, foreign words and neologisms (e.g., “ROTFL”, “C’est la vie” when spoken in a non-French voice). To customize these pronunciations, you upload an XML file with lexical entries. For example, you can customize the pronunciation of Nguyen by providing a phoneme using this XML:

Nguyen (before)
<lexeme>

<grapheme>Nguyen</grapheme>

<grapheme>nguyen</grapheme>

<grapheme>NGUYEN</grapheme>

<phoneme>"nu.jEn'</phoneme>

</lexeme>
Nguyen (after)

Brand Voice is a custom engagement where you work with the Amazon Polly team to build an Neural Text-to-Speech (NTTS) voice for the exclusive use of your organization. Brand Voice allows you to differentiate your products and applications with a unique vocal identity in a wide variety of use cases, including Amazon Connect and Alexa Skills integrations. We work with you throughout the entire process to identify the persona, identify an actor or actress and record their speech, and ultimately build and train a model to produce the voice. The voice is then made available to your AWS account ID(s).

Listen to National Australia Bank Brand Voice

Listen to Bank of New Zealand Brand Voice

If you are interested in building a Brand Voice using Polly, please reach out to your AWS Account Manager or contact us for more information.