Languages and runtime data playback

Series: About language support in Genesys Cloud and Architect

Previous suggested step: Runtime data playback overview

When you select a language in a flow, Architect returns a message if:

The default TTS engine does not support runtime data playback or TTS playback for the selected language
If no available TTS engine integration supports the language

If you attempt to process an expression in an unsupported language, then Architect returns a validation error. If you run the flow for the unsupported language, then Architect plays back blank audio.

Note: The organization's default text-to-speech (TTS) engine is Genesys TTS. To access additional third-party TTS engines in Architect, first obtain them from the Genesys AppFoundry. Then, configure them in Genesys Cloud. For more information, see About text-to-speech (TTS) engines.

In Architect, TTS engines obtained from Genesys AppFoundry work the same for both static and dynamic prompts. Dynamic prompts, such as playing back a balance at runtime, must be cast to a string before playback. For example: ToAudioTTS(ToString(Flow.AccountBalance)).

About design time validation errorsClick to expand

Find details about error messages

When an expression contains an error, Architect returns a message below the editor. To view details about the error, click the message and notice the following features:

If the error contains a function name (for example, MakeDuration), then the function name appears in blue. Click it to open the Expression Help dialog box on the specified function's focus.
The remainder of the error message appears in red. Click it to focus on the expression editor and scroll the input caret to the error location.

Configure TTS playback for languages in which TTS is not supported

Configure TTS playback in a language that Architect’s default TTS engine, Genesys TTS, or a third-party TTS engine supports. Similarly, when you configure data playback using a ToAudio function, ensure that Architect supports runtime data playback for the language.

For example, assume that you select the following supported languages in a call flow’s Settings area:

English (en-US)
Breton (br-FR)

Then, for the call flow’s initial greeting, you add a TTS greeting of “Welcome” to the audio sequence. In this case Architect returns a validation error because it does not support TTS for Breton. The TTS expression is invalid; Architect interprets the action as an attempt to play back TTS in a language that has no TTS engine.

However, you can specify an alternate Breton audio sequence to resolve the language error. In the Edit Audio Sequence dialog box, enable an alternate sequence for Breton, create an audio prompt based on the text, and add a valid audio resource. Here, Architect validates the main sequence expression because the TTS plays in English, not Breton, and Architect supports TTS and runtime data playback in English.

Alternate sequences pass to the parser as the designated language that the system checks for either runtime data playback or TTS playback. The main sequence expression passes in all flow supported languages minus any defined alternate sequence languages.

[NEXT] Was this article helpful?

Get user feedback about articles.