-
-
Main Page
-
-
TempleGuardianAI Agent (taking Data From User )
-
ALL Temples
-
ALL Temples in Grid Manner
-
ALL Temples in Grid Manner
-
Temple Details
-
Temple Chantings (Songs and Manthras)
-
Navigations To Temples
-
AI Chat Assisstence for TempleGuardianAI Users and Senior Citizens
-
Start To Assisstence
-
Voice To Voice AI Assisstence (In Main Page )
-
Temple History from AI Assisstence
Inspiration
What it does## About the Project — TempleGuardianAI
Inspiration
The idea for TempleGuardianAI was born out of a desire to help senior citizens and spiritual seekers plan meaningful and hassle-free temple trips across India.
Many elderly individuals face challenges such as:
- Researching temple history and darshan timings
- Understanding rituals and booking services
- Navigating apps or accessing guides in local languages
We envisioned a friendly, multilingual AI assistant that could act as a temple trip planner, voice-based helper, and digital devotional companion — all in one.
What I Learned
While building TempleGuardianAI, I gained hands-on experience in:
- Multi-agent AI integration (OpenAI GPT, Whisper, Transformers)
- Voice-enabled apps using
speechrecognition,gTTS, andpyttsx3 - Multilingual interaction with
langdetectand translation APIs - Google Maps, WhatsApp & calendar integration
- PDF trip plans and QR code generation using
fpdf,Pillow,qrcode - Backend services like Firebase (Firestore), dotenv, and SQL Server (
pyodbc) - Analytics & tracking for user planning and app performance
🔧 How I Built It
| Area | Tools & Technologies |
|---|---|
| Frontend | Streamlit, Lottie animations, HTML/CSS for UI |
| Language & Voice | langdetect, GoogleTranslator, Whisper, gTTS, SpeechRecognition |
| PDF + QR | fpdf, qrcode, Pillow |
| AI Core | OpenAI GPT models, Transformers |
| Sharing | WhatsApp autofill links, ICS Calendar export, Email PDF sharing |
| Backend | Firebase Firestore, .env with dotenv, SQL Server via pyodbc |
Challenges Faced
- Dependency conflicts (e.g.,
httpx,googletrans) across libraries - Managing virtual environments and package isolation
- Supporting voice + text input across Indian languages (Kannada, Tamil, Hindi)
- Balancing model performance vs. system resource use
- Designing for senior users — with large UI elements, clear steps, and voice guidance
- Seamless integration of AI output into real-world utilities (WhatsApp, UPI, Maps)
TempleGuardianAI is a heartfelt fusion of devotion and digital — a small step toward making ancient journeys accessible with modern tools.
We hope this app becomes a helpful companion for temple travelers of all ages.
How we built it
Challenges we ran into
Accomplishments that we're proud of
What we learned
What's next for TempleGuardianAI
Built With
- css3
- dotenv
- firebase
- firestudio
- fpdf
- github
- google-maps
- googlecalenderapi
- googletranslator
- gpt-4
- gtts
- html
- huggingface
- javascript
- langdetect
- lottie
- lovableai
- openai
- pillow
- pyodbc
- python
- qrcode
- speechrecognition
- sqlserver
- streamlit
- transformers
- upi
- whatsappapi
- whisper


Log in or sign up for Devpost to join the conversation.