Frequently Asked Questions

VocaLoop is a multimodal interaction layer for voice AI agents. When a call reaches a step that needs a screen — picking an option, uploading a document, confirming an address, paying — your agent triggers an interactive WhatsApp or SMS experience. The customer completes it on their phone while the call is still live, and the response flows straight back into the agent's workflow.

No. VocaLoop does not transcribe, generate speech, or run the conversation. It is infrastructure that sits alongside whatever voice stack you already run. VocaLoop is not a replacement for your agent platform — it is the piece that lets it finish the job.

Voice agents can hold a conversation end to end, but conversion usually depends on an action that cannot happen in audio. The industry default is to text a link after the call, which breaks the interaction and loses intent. VocaLoop closes that gap without ending the call.

Teams running voice AI agents where calls need a visual step to close. Common use cases include hospitality (room and rate selection), insurance (document capture and payment), field service (photo capture and dispatch confirmation), logistics (address and delivery confirmation), retail (product selection and checkout), and lending or collections (KYC documents and payment).

Your voice agent calls VocaLoop as a tool when it reaches a step that needs a screen. VocaLoop delivers an interactive WhatsApp or SMS experience to the customer's number mid-call. The agent keeps talking while the customer completes the action, and the structured response is routed back into the agent's context in real time.

No. The call stays connected throughout. The customer never leaves the conversation, so the intent built during the call is not lost.

Every step can define a fallback. VocaLoop can retry on SMS, hand control back to the agent to collect the information by voice, or park the step for post-call follow-up. Wait times and agent behaviour are configurable per step.

VocaLoop plugs into Vapi, Retell, ElevenLabs, and LiveKit out of the box, and into any custom stack that can call an external tool or webhook mid-conversation.

No. VocaLoop is exposed as a tool your agent calls. You add the tool definition and describe when to use it. Your existing prompts, flows, and telephony stay as they are.

Back into the live agent's context, and to wherever else you route it: a webhook, your CRM, your core system, or your data warehouse. The response payload is structured and typed, so it can be written straight to a record without post-processing.

Yes, for WhatsApp. Interactive business-initiated messages require an approved WhatsApp Business Account and message templates. VocaLoop helps set this up during onboarding, and you can start on SMS while WhatsApp approval is in progress.

Steps can be authored in any language your customers use, including Hindi, Marathi, Tamil, Bengali, and Hinglish. The interactive experience is rendered in the language your agent is conversing in.

No. VocaLoop never touches the audio leg. It handles only the interactive step and its response.

VocaLoop does not process or store card data. Payment steps hand off to your existing payment provider, and only the confirmation status is routed back.

No. Customer data is not used to train models.

VocaLoop spun out of DronaHQ, an enterprise AI agent platform. It runs as a separate product with its own roadmap and does not require DronaHQ to use.

Write to us at inquiries@vocaloop.ai or read the documentation for setup guides and API details.