• Atul Hooda
• Siddharth

Major portions of high-intent calls into independent Indian specialty clinics-dental, dermatology, hair transplant, fertility, and ophthalmology-go unanswered since their front-desk staff is unable to keep pace with the inbound calls. One missed call from a patient with high-intent can mean foregone revenue of 40k to 3L INR per appointment, depending on specialty. Current solutions fare differently. IVR systems cannot understand natural speech. Web forms require patient action. Rule-based chatbots break on Indian colloquialisms and accents. Developer-oriented conversation AI platforms require integration work that clinic owners simply cannot perform. Standard voice-based systems are usually designed for a monolithic English context, not the multiplicity of languages used by Indians when interacting with healthcare. In this paper, we propose the architecture, design, and early operational experience of Engageo, a missed-call recovery system piloted at five Indian specialty clinics. Our system is structured around the Decision Lifecycle (6 stages, from missed-call to verified appointment and patient reporting). Engageo supports 12 Indian languages: Hindi, Hinglish, English, Punjabi, Bengali, Tamil, Telugu, Kannada, Malayalam, Gujarati, Marathi, Odia, and allows patient-driven language-switching at turn boundaries. We have 5 main contributions: (1) The Decision Lifecycle framework for both engineering and human communication. (2) A multilingual conversation architecture across 12 Indian languages. (3) The Script Coherence Rule, a prompt-level constraint to address a TTS failure specific to Indian multilingual systems: text-to-speech with a mixed-script output results in unintelligible audio when multiple scripts are concatenated together (e.g. Devanagari concatenated with Gurmukhi, Devanagari with Tamil). (4) A 5-stage messaging lifecycle: Confirmation, Reminders, Follow-up, Missed Call follow-up, and No-Show recovery. (5) Engineering patterns for systems like this, such as race condition handling during calendar booking, atomic guards against redundant sends, and Indian phone number normalization. This paper is a systems-and-design contribution. We share qualitative results from roughly 50 in-depth interviews with Indian specialty clinic owners, and from early deployments at our five pilot clinics. We are performing a quantitative outcome study with these clinics and plan to report these results separately.

Voice AI, Multilingual Conversational Agents, Indian Healthcare, Missed-Call Recovery, WhatsApp Automation, Appointment Scheduling, Script Coherence, Code-Switching, Speech Recognition, Healthcare Informatics, Specialty Clinics, Patient Acquisition

"Engageo: A Multilingual Voice and Messaging System for Missed-Call Recovery in Indian Specialty Clinics", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.13, Issue 5, page no.g639-g648, May-2026, Available :http://www.jetir.org/papers/JETIR2605676.pdf

"Engageo: A Multilingual Voice and Messaging System for Missed-Call Recovery in Indian Specialty Clinics", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.13, Issue 5, page no. ppg639-g648, May-2026, Available at : http://www.jetir.org/papers/JETIR2605676.pdf