Configuration Reference

Prompt Module Properties

JSON type object

Location Configuration root

JSON property Prompt Module object JSON type object

These properties define reusable prompt content.

Behavior and notes

For directly specified content, Andamento resolves playback flexibly: audioResource first; if that resource is unavailable, text; then mapped fallback text from the Audio Resource Text Map; then textResource.
Prompt Modules are reusable. Conversation modules and other Prompt Modules reference them by ID, so playback content can be changed without restructuring the Conversation Flow.
Runtime-resolvable prompt content is useful for application-generated responses. The text property supports both a complete runtime field and embedded runtime fields such as "Hello $[CollectedName]".
sequence plays all referenced prompts during one playback operation. serial is specifically useful with Interactive Capture retry/event prompts: successive occurrences advance through the prompt list, and after the final prompt is reached the final prompt is reused.
resourceLanguage controls resource lookup, not speech-synthesis voice selection. Voice selection is controlled separately through speechSynthesisVoice and related synthesis settings.

Developer guidance

A Prompt Module is a reusable prompt definition identified by id. It can specify an audio resource, inline text, or a text resource, along with playback mode, language, speech synthesis settings, and playback parameters.

Example in context

Illustrative JSON excerpt (not a complete configuration document):

{
"id": "prompt.welcome",
"playbackMode": "single",
"text": "Welcome to SpeechTiles."
}

This is a Prompt Module object, not a complete Prompt Module configuration file.

Properties

Jump to a property:

id · playbackMode · prompts · audioResource · text · textResource · isSSML · autoDetectSSML · autoSelectVoiceForLanguage · resourceLanguage · speechSynthesisEngine · speechSynthesisVoice · audioPlaybackParams

`id`

stringRequired

Specifies the unique identifier of the Audio Prompt Module. The identifier is used by conversations and other prompt modules to reference this module.

Runtime fields: Not supported — The module identifier is structural and cannot be supplied through a runtime field.

Prompt identifier

"id": "prompt.welcome"

`playbackMode`

stringRequired

Specifies how the Audio Prompt Module selects and plays its prompt content.

Allowed values

single — Plays prompt content specified directly using audioResource, text, or textResource.
select — Randomly selects and plays one Audio Prompt Module referenced by prompts.
sequence — Plays the Audio Prompt Modules referenced by prompts as a sequence during a single playback operation.
serial — Used with interactiveCapture. Successive occurrences of an event select successive Audio Prompt Modules from prompts, such as the first timeout prompt for the first timeout and the second timeout prompt for the second timeout.

Runtime fields: Supported — To supply the value dynamically, specify a runtime field as the complete property value using the $[FieldName] syntax.

Use sequence when several prompts should play during the same operation. Use serial for successive Interactive Capture event occurrences rather than as a general multi-prompt sequence.

`prompts`

arrayOptional

References reusable Prompt Modules used by select, sequence, and serial playback modes.

Prompt references

"prompts": [
  { "promptID": "prompt.first" },
  { "promptID": "prompt.second" }
]

`audioResource`

stringOptional

Specifies the audio resource to play. When defined, Andamento attempts this resource before text alternatives. If the resource cannot be found or played, Andamento can fall back to the available text sources.

Runtime fields: Supported — To supply the value dynamically, specify a runtime field as the complete property value using the $[FieldName] syntax.

`text`

stringOptional

Text to synthesize for playback. It can be literal, supplied entirely by a runtime field, or contain embedded runtime fields. It is considered after audioResource and before mapped fallback text and textResource.

Runtime fields: Supported — The complete property value may be a runtime field using $[FieldName]. Embedded runtime fields are also supported within otherwise literal text.

Runtime field

"text": "$[AssignedText]"

Embedded runtime field

"text": "Hello $[CollectedName], welcome back."

`textResource`

stringOptional

Specifies a resource containing text to synthesize. For directly specified prompt content, textResource is the final source in the fallback chain after audioResource, text, and mapped fallback text.

Runtime fields: Supported — The resource identifier may be supplied dynamically as the complete property value. No embedded interpolation behavior is documented for the contents of the referenced text resource.

`isSSML`

booleanOptionalDefault: False

Specifies whether the prompt text is explicitly SSML. When true, Andamento treats the text as SSML regardless of autoDetectSSML.

Runtime fields: Supported — When isSSML is false and autoDetectSSML is true, Andamento auto-detects whether the text is SSML.

`autoDetectSSML`

booleanOptionalDefault: True

Specifies whether Andamento automatically detects SSML when isSSML is false. To force text to be treated as plain text, set both isSSML and autoDetectSSML to false.

Runtime fields: Supported

`autoSelectVoiceForLanguage`

booleanOptionalDefault: False

Specifies whether Andamento automatically detects the language of the text and selects an appropriate speech-synthesis voice for the detected language.

Runtime fields: Supported

`resourceLanguage`

stringOptional

Specifies the language used when resolving language-specific prompt resources such as audioResource and textResource. Andamento first searches the corresponding language folder and, if the resource is not found there, falls back to the parent resource folder one level above the language folder.

Runtime fields: Supported

`speechSynthesisEngine`

stringOptionalDefault: Apple

Specifies the speech-synthesis engine to use for the prompt. The property supports selecting among speech-synthesis engines as additional engines become available. Currently Andamento supports the Apple speech-synthesis engine. If not specified, Apple is selected.

Runtime fields: Supported

`speechSynthesisVoice`

objectOptional

Specifies criteria used to select the speech-synthesis voice for the prompt.

`audioPlaybackParams`

objectOptional

Specifies optional playback controls, inserted speech-synthesis silence, and synthesized-speech recording behavior for the prompt.

Examples

Single synthesized prompt

[
{
"id": "prompt.welcome",
"playbackMode": "single",
"text": "Hello $[CollectedName], welcome back."
}
]

Sequence of prompt modules

[
{
"id": "prompt.sequence",
"playbackMode": "sequence",
"prompts": [
{ "promptID": "prompt.first" },
{ "promptID": "prompt.second" },
{ "promptID": "prompt.third" }
]
}
]