The Ask step in Edesy IVR speaks a question, listens, and turns the caller's spoken answer into text. You name the answer, set how long to wait, re-ask when the caller is silent and list expected words. Answers show in call logs and status callbacks. Menus still use the keypad.
₹20 Free Credit, No Card
₹1.50 per Minute
Visual Flow Builder
Speak a question, listen, save the answer, move on
Drop it after your greeting. Give it a question key such as location; that key names the answer everywhere else.
Type the question the caller hears, in Devanagari, Roman script or English. Preview the audio in the builder before any call.
Speech is recognised in English (India), English (US) or Hindi. Set it on the step, or leave it on a Config step set to one of those.
Set how long to wait for an answer, how long a pause ends it, how many times to re-ask, and the words callers are likely to say.
Answered goes to the next question or step. No answer goes wherever you want silent callers to go, such as reception.
Use {{answer.location}} in an HTTP Request or Email step to post the text to your CRM or inbox, then test call and publish.
| Setting | Range or default | What it does |
|---|---|---|
| Question key | Required. Lowercase letters, digits and _ |
Names the answer in call logs, callbacks and {{answer.<key>}} tokens |
| Question | Required | The prompt spoken to the caller. Supports {{var.name}} from the outbound call API, with audio preview |
| Language | English (India), English (US) or Hindi, or the flow default | Used to speak the question and to transcribe the answer |
| Wait for an answer | 1 to 30 seconds, default 8 | How long to wait for the caller to start speaking |
| Pause that ends the answer | 1 to 5 seconds, default 2 | How long the caller can pause mid-sentence. Raise it for open questions, lower it for one-word answers |
| Re-ask if silent | 0 to 2, default 1 | Extra attempts before the call follows No answer |
| Re-ask prompt | Optional | Spoken instead of the question on a re-ask, for example "Sorry, could you say that again?" |
| Expected words | Optional, comma-separated | Names or terms callers are likely to say, such as city or product names, to improve recognition |
Exits: Answered and No answer. If No answer is left unconnected, a silent caller simply moves on down Answered, so a survey keeps going.
Languages: the Ask step listens in English (India), English (US) or Hindi. Set the language on the step, or on a Config step after Start that sets the voice for the whole call. Check recognition with a test call before going live.
location: "आप किस शहर से बात कर रहे हैं?" Expected words: Delhi, Lucknow, Jaipur.reason: "कृपया बताइए, आप किस बारे में कॉल कर रहे हैं?" Pause that ends the answer: 3 seconds.{
"caller": "{{caller}}",
"city": "{{answer.location}}",
"reason": "{{answer.reason}}",
"call_sid": "{{call_sid}}"
}
{{var.*}} tokens only; answers are for HTTP Request and Email steps, logs and callbacks.Capture the caller's city before routing them, and send it to the team that picks up.
Let callers say why they are calling in their own words, logged against the call.
Follow a 1 to 5 keypad rating with "tell us why", captured as text.
Ask for a name or order reference on an outbound call and post it to your system.
Share your use case and volume. We reply within one working day with a setup plan and pricing.
Resources to help you evaluate and implement
Build a flow with an Ask step, preview the audio and place a test call. ₹20 free credit, no card.