Facebook Instagram Youtube Linkedin Twitter

Our Response to COVID-19

A Human-Centered AI Framework for Multilingual Rural Field Reporting, Qualitative Insight, and Community Support_New

Human-Centered AI Framework

AI That Understands Communities, Context, and Human Stories

A multilingual rural field reporting framework designed to support qualitative insight generation, participatory research, and community-centered decision making through ethical and human-assisted Artificial Intelligence.

Multilingual

Supports mixed-language and local dialect reporting

Human-AI

Combines AI processing with contextual human validation

Inclusive

Designed for rural and low-resource environments

Qualitative

Focused on lived experiences and grassroots insight

A Human-Centered AI Framework for Multilingual Community Development & Support Field Reporting, Qualitative Insight, and Community Support

(Sruthi · Sakhi · Viveka AI · AI-Facilitator)

“AI that listens first — built with women’s collectives, for the realities of multilingual, low-literacy community life.”
Combination: Applied Research + Product Development + Pilot Implementation + Evaluation. Includes participatory co-design, naturalistic field deployment, and longitudinal implementation-science evaluation.
Initiated 2023 (Sruthi field pilot 2024; Viveka AI pipeline and ICGT 2025 publication; ongoing portfolio integration through 2026).
Context-Aware AI for Social Good
Women’s Empowerment
Implementation Science
Participatory AI
Human-Centered Computing
Frontline community workers; field practitioners and emerging social workers; program coordinators, supervisors, and researchers; tech enthusiasts motivated to make real social impact; institutional decision-makers in rural development; women in SHGs or cooperatives; and Gender Point Persons (GPPs).
  • Technologies designed for low digital literacy contexts, with simple and user-friendly interfaces that are practical and accessible in real-world use.
  • Conversational platforms such as WhatsApp and Telegram, already embedded in community communication ecosystems.
  • Multilingual AI and speech systems including LLMs, ASR, and TTS tools supporting code-switching across major Indian languages.

Most AI is built for scale, speed, and engagement. This project asks a different question: what happens when AI is built for listening — to women’s collectives, field practitioners, and multilingual communities whose realities are often missed by standard systems?

The portfolio — Sruthi, Sakhi, Viveka AI, and an AI-Facilitator — combines conversational AI and multilingual speech tools that help community support teams and researchers capture, interpret, and act on community knowledge across languages and literacy levels.

Grounded in the AWESOME framework for women’s empowerment and aligned with UNESCO’s Recommendation on the Ethics of AI and NITI Aayog’s Responsible AI framework, the work prioritises transparency, human oversight, and community accountability from the very beginning.

Project Background & Motivation

Designing Human-Centered AI for Women-Led Rural Community Support Systems

Problem / Need

Why this work matters

Picture a community field worker at the end of a long day — she has just witnessed something important: a moment of trust in a Self-Help Group meeting, a woman cautiously speaking about safety, or an aspiration that took months to surface.

She opens WhatsApp and types a few lines. No one responds. Her supervisor is stretched across dozens of others. The reporting system asks her to fill out structured forms with no space for what she actually observed. By the next morning, the moment is gone.

This is the challenge the project addresses. Community support programs generate rich and real-time field knowledge, yet much of it disappears into informal voice notes, fragmented conversations, and reporting structures that cannot capture lived experiences.

Women’s voices — especially around safety, dignity, opportunity, and aspiration — rarely travel from the community to the systems designed to respond to them.

The project explores whether AI, designed differently and governed responsibly, can help bridge this gap while preserving human context, trust, and ethical oversight.

Goals & Objectives

Building a field-to-insight ecosystem for community support

System Goal

Build and evaluate an integrated field-to-insight loop linking conversational support, multilingual AI interpretation, and actionable follow-up for facilitators in SHG and PRA settings.

Social Goal

Strengthen women-centered reporting, participation, reflection, and support across livelihood, education, health, safety, social participation, and environmental dimensions.

Research Goal

Develop longitudinal evidence on the acceptability, feasibility, adoption, fidelity, and sustainability of multilingual voice-first AI systems in low-resource rural settings.

Governance Goal

Create reusable accessibility, consent, contestability, and oversight frameworks co-developed with women’s collectives and field teams.

Innovation & USP

Human-centered innovation designed with communities, not for them

Peer-Mode, Not Surveillance

Sruthi is designed as a trusted peer within practitioner groups rather than as a monitoring layer for management systems.

Community-Owned Insight

Field knowledge flows back toward communities as usable and actionable insight instead of becoming extractive institutional data.

Multilingual & Voice-First

Every component supports voice-first and code-switched interaction across Indian languages, making accessibility central rather than optional.

Human Oversight by Design

AI-assisted interpretation always includes facilitator validation to preserve contextual understanding and reduce harmful misinterpretation.

Theory Before Technology

Every feature is mapped to women’s empowerment outcomes and Theory of Change frameworks before development begins.

Governance Before Scale

Consent, contestability, anonymisation, and the right to withdraw are built into the system before deployment and scale-up.

UN Sustainable Development Goals

Aligned with SDG 5, SDG 10, SDG 4, SDG 3, SDG 16, and SDG 17 focusing on gender equality, inclusion, education, health, and partnerships.

SDG 5
Gender Equality
SDG 10
Reduced Inequalities
SDG 4
Quality Education
SDG 3
Good Health
SDG 16
Peace & Justice
SDG 17
Partnerships
Human-Centered AI Framework

Designing AI Systems That Listen to Communities, Context, and Women’s Voices

A multilingual, voice-first framework for rural field reporting, qualitative insight generation, and community-centered support systems designed around trust, participation, and ethical governance.

Some of the most important field insights disappear before anyone can respond.

Picture a community field worker at the end of a long day — she has just witnessed something that matters: a moment of trust in a Self-Help Group meeting, a woman cautiously speaking about safety, or an aspiration that took months to surface.

She opens WhatsApp and types a few lines. No one responds. The reporting system asks her to fill out structured forms with no space for what she actually observed. By the next morning, the moment is gone.

Community support programs generate rich and real-time field knowledge, yet much of it disappears into fragmented conversations, informal voice notes, and systems unable to capture lived experiences.

“ Women’s voices are often collected as data — but rarely returned as actionable support. ”
Goals & Objectives

Building a field-to-insight ecosystem for community support

The framework combines conversational support, multilingual AI interpretation, and human validation to strengthen participation, reflection, and women-centered support systems in low-resource rural settings.

01

System Goal

Build and evaluate integrated field-to-insight loops linking reflective support, multilingual AI interpretation, and actionable follow-up for facilitators in SHG and PRA environments.

02

Social Goal

Strengthen women-centered participation, reporting, reflection, and support across livelihood, education, health, safety, social participation, and environmental dimensions.

03

Research Goal

Develop implementation-science evidence on feasibility, fidelity, sustainability, and adoption of multilingual voice-first AI systems in low-resource field settings.

04

Governance Goal

Create reusable accessibility, consent, contestability, and oversight guardrails co-developed with women’s collectives and field practitioners.

Innovation & Principles

Human-centered innovation designed with communities, not for them

01

Peer-Mode, Not Surveillance

Sruthi is designed as a trusted peer within practitioner groups rather than as a monitoring layer for management systems. This is an ethical commitment, not simply a technical feature.

02

Multilingual & Voice-First by Design

Every component supports voice-first and code-switched interaction across Indian languages, ensuring accessibility for communities often excluded by English-first digital systems.

03

Human Oversight Remains Central

AI interpretation always includes facilitator review and contextual validation to preserve trust, reduce harmful misinterpretation, and maintain community accountability.

Alignment & SDGs

Aligned with global, national, and community-centered priorities

The framework aligns with international sustainable development priorities, responsible AI governance principles, and women-centered rural development initiatives.

SDG 5
Gender Equality
SDG 10
Reduced Inequalities
SDG 4
Quality Education
SDG 3
Good Health
SDG 16
Peace & Justice
SDG 17
Partnerships

The future of AI for social good is not about replacing communities. It is about building systems capable of listening.

Human-centered AI must strengthen participation, preserve dignity, support multilingual realities, and return insight back to the communities from which it emerges.

TYPE AND PRESS "ENTER" TO SEARCH