Enhanced Text-to-Speech

Enhanced Text-to-Speech converts your written articles into natural, human-like audio using voices from third-party text-to-speech providers. Once activated, audio is automatically generated for new publications, allowing readers to listen to articles instead of reading them. This makes your content more accessible and offers an additional way for readers to enjoy it.

What is it used for?

Enhanced Text-to-Speech uses voices from supported text-to-speech providers instead of the standard voices built into browsers and mobile devices. These voices sound more natural, with improved pronunciation, rhythm, and intonation, creating a better listening experience.

You might choose Enhanced Text-to-Speech if you want to:

• Make your content more accessible.
• Give readers the option to listen to articles instead of reading them.
• Improve audio quality, particularly for languages where standard device voices are limited

How is it implemented?

To activate Enhanced Text-to-Speech, you first need an account with one of our supported text-to-speech providers. Once you have created your account and obtained the required credentials, send them to hello@prenly.com and we will configure the integration for your publication.

You can compare available voices and find setup instructions for the supported providers on their respective websites. Prenly currently supports:

• Microsoft Azure Voice Gallery
• Google Cloud Text-to-Speech
• Amazon Polly Voices
• ElevenLabs
• Narakeet

The exact credentials depend on the provider you choose. In general, you will need to provide:

• Voice (for example, en-US-JennyNeural)
• Speech key, API key, or JSON key file (depending on the provider)
• Region
• Language
• Voice gender

After the integration has been configured, continue uploading your publications to Webarch as usual. Audio is then generated, imported, and made available automatically.

Be aware: Only publications uploaded after the integration has been activated will include enhanced audio. Existing publications are not processed automatically.

Technical details

The quality of generated audio depends on the selected text-to-speech provider and voice.

Some providers calculate character usage based not only on the article text but also on additional markup used for language and voice settings. This may affect the number of billable characters.

If you would like to activate Enhanced Text-to-Speech for your publication, have questions about the setup, or would like pricing information, please contact hello@prenly.com.