Skip to main content

← Projects · agriculture

AI-assisted Farmer Call Center

Automated voice-response system letting farmers report issues via phone, SMS, or web and receive AI-routed answers.

The problem statement asked for a voice-first interface so smallholder farmers without smartphones or stable data could report issues (pest sightings, crop disease, weather questions) by calling, texting, or visiting a simple web form. The system replies in the farmer’s language with appropriate guidance, escalating to human experts when the model is uncertain.

Solution sketch:

  • Inbound layer. Twilio (or local PSTN gateway) accepts calls and SMS; web form mirrored.
  • Speech-to-text. Local-language ASR, fallback to English.
  • Reasoning layer. Classification and retrieval over an agricultural-extension knowledge base.
  • Outbound layer. TTS reply by phone; SMS or notification on web.

Outcome: AGRI12 winners. See the UNESCO India-Africa 2022 award page for the full team roster, prize details, and source citation.