Objective
This article shows you how to add and configure the Gather Input on Call widget so your Studio flow plays a prompt and then captures what the caller presses on the keypad or says out loud — the core of any phone menu or IVR.
Product
Studio, Programmable Voice
Environment
Twilio Console
User Account Permission/Role(s) Required
Editing a flow requires the Studio flows create and update permissions on the account.
Procedure
- In the Twilio Console, select Orchestration in the left sidebar.
- Select Studio, then Flows.
- Select the flow you want to edit to open the Editor tab.
- Open the widget library and drag Gather Input on Call onto the canvas.
- Select the new widget to open its configuration panel.
- In Say or play message, choose how the prompt is delivered:
- Say a message — enter the words in Text to say, then choose a Message voice and a Language, such as English (US).
- Play a message — enter the file's public address in URL of the audio file, and set Number of loops if you want it repeated.
- Set Stop gathering on keypress? if the prompt should be interrupted as soon as the caller presses a key.
- To accept spoken answers, configure the speech settings:
- Speech model: choose the model that matches your call type, such as Phone Call, Conversations, Numbers & Commands, or Manual.
- Speech recognition language: the language the caller will speak.
- Speech recognition hints: a comma-separated list of words or phrases to bias recognition toward.
- Speech input timeout: how long to wait for the caller to finish speaking.
- Profanity filter: whether recognized profanity is masked.
- Select Save.
- Open the Transitions tab and choose where the flow goes next for If user pressed keys, If user said something, and If no input.
- Select Publish to make the change live.
Additional Information
- Speech recognition and keypad input can both be active on the same widget; the transition that fires tells you which the caller used.
- Number of loops and URL of the audio file apply only in play mode; Text to say, Message voice, and Language apply only in say mode.
- The captured value is available to later widgets in the flow, so pair this widget with Split Based On... to route on the answer.
This content was generated by AI and reviewed, edited, and verified by a human.