Grammatical gender in audio sequences
Architect provides gender specification for supported languages in audio playback of numerical prompts. When a number must be played back in a specific gender context, Architect contains parameters that allow the flow author to specify how TTS plays the content back to the caller at runtime. Valid gender categories are:
- Masculine: Indicates a masculine gender value
- Feminine: Indicates a feminine gender value
- Common: Indicates the content may be of either masculine or feminine gender value
- Neuter: Indicates gender-neutral or inanimate value with no masculine or feminine association
- Articles: Indicates that the article must agree with the selected gender
Configure these settings for data values when you create an audio sequence. In the Edit Audio Sequence dialog box, when you select a data element for which grammatical gender applies, the dialog box changes to include an additional drop-down list. The following table describes situations that apply to gender classification settings:
|Audio Sequence language||Gender classification drop-down list||Architect behavior|
|Main sequence||Gender classification drop-down list appears, whether the main sequence language requires it or not||If gender classifications are not required for the Main Sequence language, Architect ignores configured settings.|
|Alternate sequence for a language that requires gender classification||Gender classification drop-down list appears||Architect follows configured gender classification settings for the language’s alternate sequence|
|Alternate sequence for a language that does not require gender classification||Gender classification drop-down list does not appear||Not applicable|
In an expression, a flow author can specify gender or case as appropriate for a language using enum values from the built-in Language variables, such as Language.Gender or Language.Case. Function overloads that support language context can be found in Expression Help.
In the audio sequence editor, if language context settings that apply to a particular piece of data playback are available, they appear as an option in the Edit Audio Sequence dialog box. For an example, click to expand the following section:
In this example, when configuring the Data element for the Spanish (es-US) value, the gender drop-down list appears.
If you select the Main Sequence which, in this case is English (en-US), the gender drop-down list for the same Data element value appears, but because gender settings are not required for this language, Architect ignores any configured settings.
Finally, if you select to enable an alternate sequence for a language that does not require gender classifications (other than the Main Sequence), this drop-down list will not appear.