For each supported language in a call flow, you can choose a TTS engine and voice at call flow runtime to read text-to-speech playback. You can select a specific engine and voice, which Architect uses for that language every time the flow runs. 

You can also select Default, which is the organization’s default engine and voice configured for that language in Genesys Cloud’s  Admin > Account Settings > Organization Settings. For more information, see Choose a default TTS engine. If you select Default, Architect displays the engine and voice that it uses, based on the organization settings.

You can also choose not to use a TTS engine for a particular language. By selecting None, any attempts to use TTS for that language within the flow generates a validation error.

When you select a third-party TTS engine that you install as an integration, be aware of third-party TTS engine limitations and behaviors. Because third-party configuration may change, Architect does not provide differences between TTS engines. However, if maximum length limitations exist, click Info for more details.

When you publish a flow, Architect locks the TTS engine and the voices you set on the supported languages, and uses them at runtime. If the organization’s default TTS engine or voice changes, and you want existing flows that use this option to update, you must republish the flow. Otherwise, the flow uses the default TTS engine and voice selected upon the original publish. To see the TTS engines and voices in flows, see Search for flows that use a specific TTS engine.