Settings & Diagnostics
Local device identity, live system diagnostics, Indic voice engine settings, and preferences for notifications and knowledge graph visualization.
Device Identity & Citizen Profile
Device Node Fingerprint
dev-local-user
Encrypted per-browser key ensures anonymous zero-leakage access.
Cases Registered
Appeals Drafted
Evidence Documents
Multilingual Indic Voice Engine (iNovel Architecture)
Plug-and-play STT → LLM → TTS full-duplex orchestrator with dynamic mid-call language switching and low-latency turn taking.
Saaras ASR acoustic conformer & Bulbul neural TTS models for Indian scripts, code-mixing (Tanglish & Hinglish), and dialects.
Filler word removal, spontaneous speech self-correction, auto-punctuation, and statutory entity normalization (IRDAI, RERA, DGCA).
Supported Indic Languages & Scripts
English
English (India)
தமிழ்
Tamil
हिन्दी
Hindi
తెలుగు
Telugu
ಕನ್ನಡ
Kannada
മലയാളം
Malayalam
বাংলা
Bengali
मराठी
Marathi
ગુજરાતી
Gujarati
ਪੰਜਾਬੀ
Punjabi
ଓଡ଼ିଆ
Odia
اردو
Urdu
Tanglish (தமிழ்)
Tanglish
Hinglish (हिन्दी)
Hinglish
System Diagnostics
Overall Backend Status
Connecting...
Env: development
Vector Engine
Backend: jsonl
0 indexed knowledge points
Knowledge Graph
Backend: jsonl
0 citizen precedent edges
Indic Voice Engine
Sarvam + Bolna + Voice
30 languages · WebSpeech/TTS
Interface Language · மொழி
Navigation and labels switch instantly and are remembered on this device.