NetSapiens icon

NetSapiens

Interact with the NetSapiens API

Actions477

Media/Music on Hold/User → Create a New MOH for Domain from TTS

AI-generated

Summary

Creates a user-level music-on-hold (MOH) item for a domain by generating audio from the supplied text-to-speech script. Choose the domain and user, then provide the script; voice settings and the optional MOH index control the generated audio.

Inputs

  • Domain (required) — Select a domain or enter its name.
  • User (required) — Select a user in the chosen domain or enter the user ID.
  • Synchronous (required) — Yes (default) waits for a 200 response; No requests an asynchronous 202 acknowledgement.
  • Script (required) — Text-to-speech content or a description of the audio to generate.
  • Index — Optional numeric MOH position; 0 is the intro greeting. Leave empty for automatic assignment.
  • Voice Language — TTS voice language code, defaulting to en-US.
  • Voice ID — TTS voice identifier, defaulting to en-US-Wavenet-C.
  • Refresh Dropdown Cache — When enabled, bypasses cached dropdown values and fetches fresh options.

Output shape

The API response for the MOH creation request.

The response may be a completed 200 response or, when asynchronous mode is selected, a 202 acknowledgement. The exact response fields are not established by the supplied source view.

Examples

Example 1: Create a user-level MOH recording for a domain

Select a domain and user, enter the spoken script, and optionally specify an index, voice language, and voice ID.

Links

Discussion