Choose or change voices
Choose or change the AI voices that read your content aloud.
This guide explains how to choose and change AI voices when creating audio with BeyondWords.
You can change these default voices in the 'Settings (General)' section of your project dashboard (please note this will also affect the default voices for the TTS Editor). Press the play icon next to any voice option to hear a preview.
Audio created using our API will use a maximum of two voices by default: one to read the title, and one to read the body content. Default voices can be changed in the 'Settings (General)' section of your project dashboard (please note this will also affect the default voices for the TTS Editor). Press the play icon next to any voice option to hear a preview.
When using HTML, you can manually set voices at a title or paragraph level by adding a
data-speechkit-voice=*data attribute to the
*is the voice ID from our list of voices. You can also override the default voice for any paragraph by editing your audio in the TTS Editor.
When creating or editing audio in the Editor, you can change the voice associated with any title or paragraph by selecting the voice name then selecting a new voice from the menu. Press the play icon next to any voice option to hear a preview.
The best AI voice for you will depend on your brand, content, and audience.
Generally, you want a lifelike voice that listeners enjoy listening to and find easy to understand. It should also reflect the nature of your content and brand. We recommend listening to previews and creating extended demos in our TTS Editor before making your final decision.
When choosing a voice from a dropdown list (in settings or the TTS Editor), you can press the play button next to any option to hear a short preview.
New voice settings will apply to newly processed audio only. When default voices are changed, pre-existing audios will not be automatically updated or updated when reprocessed. Instead, you will need to edit voice selections manually using the API or TTS Editor, or recreate the audio.