Fair Forward - Open Data & Use Cases

The AI model based on this dataset enables efficient and cost-effective remote monitoring of biomass changes. This is crucial for assessing reforestation success and detecting forest degradation due to cocoa farming. Also it reduces the need for extensive and expensive on-the-ground surveys. The dataset and the models developed from it aim to predict biomass levels in shaded regions of Côte d'Ivoire using a combination of GEDI, Sentinel-2 satellite imagery, and ground truth biomass measurements.

This project focuses on creating validated, open-source High Carbon Stock (HCS) forest maps for Indonesia using an AI-driven approach. It adopts the HCS classification system to identify forests suitable for conservation versus sustainable development: • High, Medium, and Low-Density Forests (Conservation Priority): Forests with closed or patchy canopies, rich biodiversity, and carbon storage. • Young Regenerating Forests: Regrowing forests that are transitioning into higher carbon stock potential. • Scrub and Open Land: Areas with minimal carbon stock suitable for sustainable development. The maps integrate field data, satellite imagery, and AI to assist policymakers, NGOs, and businesses in making data-driven decisions for sustainable land-use planning and conservation. Stakeholders can use the outputs for spatial planning, reducing greenhouse gas emissions, and implementing "no deforestation" commitments. The High Carbon Stock Approach (HCSA) project was nominated for the prestigious UN SDG Game Changer award – Category 2: Planet – Innovation for our climate and environment “Artificial Intelligence Meets Jungle/ Forest Forward.

FAIR Forward and Prosa.ai collected AI training data and trained models for three digitally underrepresented languages of Indonesia: Balinese, Bugis and Minangkabau. Existing works on underrepresented languages mostly focus on machine translation and simple language understanding tasks, such as sentiment analysis and topic classification. More complex tasks such as open-domain dialogue and dialogue summarization, are still left behind for these underrepresented languages which leads to a poor evaluation suite for assessing the capability of large language models (LMs) in these underrepresented languages. Mitigating this limitation, we introduce NusaDialogue, the first underrepresented languages datasets consisting of manually annotated dialogue along with its summary. NusaDialogue covers 3 underrepresented languages under the Malayo-Polynesian languages group, i.e., Minangkabau (min), Balinese (ban), and Buginese (bug), each consisting of 10,000 dialogue and summary pairs. The dialogues in NusaDialogue encompass numerous topics including culture, occupation, politics, science, history, news, sports, religion, etc

By collecting more than 1000 hours of AI voice data (by 04/2025), this effort created the largest open-source voice dataset of diverse Swahili speakers for speech recognition (speech-to-text). You can use this resource to build AI systems understanding spoken Swahili e.g. in cellphones . This facilitates access to information in mother tongue, including by illiterate persons and persons in rural settings.

Developing products that can be replicated through four initiatives: 1. DynAG: Focused on rice, wheat, and maize, DynAG's platform uses a RAG pipeline powered by LLMs like the GPT suite to provide farmers with advisory services. Accessible via chatbots, Android and Jio-based mobile apps, IVR, and SMS, it ensures wide reach. With machine translation and speech recognition, the platform supports communication in English and Hindi while utilizing CGIAR's data backbone for accurate, data-driven insights. 2. Digital Green: Digital Green's platform supports agricultural value chains like dairy, coffee, and wheat in Bihar and Kenya. It integrates a RAG pipeline, using LLMs under evaluation, and offers services through an app, Telegram, and IVR. Features like machine translation, speech technology, image recognition, and weather data enable localized insights. The platform supports multiple languages, including Hindi, Swahili, Gikuyu, and Marathi, ensuring flexibility across regions. 3. Viamo and Partners: This platform targets farmers in Kenya and Bihar, focusing on beans and wheat, respectively. Using IVR, SMS, and WhatsApp, the platform is built on a RAG pipeline powered by LLMs, with potential for intent and entity recognition. Machine translation and speech-enabled IVR support English, Hindi, and Swahili, making it accessible across regions. The initiative is part of a broader plan to enhance voice technology for agricultural advisory. 4. Tech for Her: Tech for Her empowers women farmers in Bihar and Kenya, focusing on tomatoes and cattle. Accessible via IVR, WhatsApp, and a mobile app, the platform uses LLM-powered intent and entity recognition for personalized chatbot interactions and speech-enabled IVR for voice support. AI-based video generation further enhances engagement. Available in English, Hindi, and Swahili, it caters to the specific needs of women in agriculture.

This effort created the largest open-source voice dataset of diverse Kinyarwanda speakers for speech recognition (speech-to-text). It collected more than 2300 hours of AI voice data in Kinyarwanda (by 04/2025). You can use this resource to build AI systems understanding spoken Kinyarwanda e.g. in cellphones . This facilitates access to information in mother tongue, including by illiterate persons and in rural settings.

Development and implementation of a training program to enable practitioners in the field of Earth Observation in South Africa to use machine learning. Field data collected on small holder agricultural farm land (crops) with raw datasets + applied use cases of classification of land use cover using ML (based on individual projects presented)

Investigates methods for detecting and assessing stress in tomato plants using ASD measurements and drone data, focusing on Project Munei Holding Investment in Ha-Mphaila, Limpopo

Investigates the use of Sentinel-2 satellite imagery and a random forest (RF) machine learning algorithm to estimate the Leaf Area Index (LAI) of tomato crops in the Ha-Mphaila irrigation scheme, Limpopo Province, South Africa. The research aimed to determine key wavebands and vegetation indices for LAI estimation, compare the performance of RF models with stepwise multiple linear regression (SMLR) models, and map the spatial distribution of LAI.

Investigates the use of remote sensing and machine learning to characterize maize stress in the Limpopo Province, South Africa. The study concludes that integrating field spectra and remotely sensed reflectance provides valuable information for maize stress investigation, supporting the development of mitigation and prevention strategies to ensure food security.

Assesses the capabilities of Earth observation and machine learning algorithms, specifically Random Forest and Support Vector Machines, in detecting maize disease stress using drone data. The study was conducted on a small-scale farm in Limpopo, South Africa, focusing on healthy maize, maize with streak virus, and maize with phosphorus deficiency.

Drone high-resolution images were used with a semi-automated random forest (RF) classifier algorithm in Google Earth Engine to classify bare soil, weeds, and tomatoes. The findings can serve as a reference for agricultural researchers and advise farmers on crop management, potentially reducing costs and herbicide use.

Explores using the Random Forest Regression (RFR) machine learning algorithm with Sentinel-2 and drone imagery to estimate relative chlorophyll values in tomato and maize crops within South Africa's Ha-Mphaila farming area. The research concludes that both Sentinel-2 imagery (10m and 20m spatial resolution) and high-resolution drone images are valuable for rapidly assessing crop chlorophyll content across significant areas, offering useful insights for precision agriculture.

Evaluates the performance of machine learning algorithms for estimating chlorophyll content in tomatoes using Sentinel-2 satellite data.

Explores the integration of ground-based vegetation parameters, thermal infrared data from handheld cameras, and UAV multispectral data to map crop canopy temperature at high resolution. The research, conducted at the Ha-Mphaila irrigation scheme in Limpopo, South Africa, focused on tomato crops.

Investigates how remote sensing and machine learning can be used to improve Agricultural Index Insurance (AII) for smallholder farmers in South Africa, who often lack access to affordable insurance. Key recommendations include supporting further research and development in these areas and implementing educational campaigns to build farmer trust and awareness of index insurance benefits.

Explores the use of Sentinel-1 and Sentinel-2 satellite data for crop type mapping in smallholder farming areas. The study focuses on improving classification accuracy by identifying optimal spectral bands and evaluating machine learning algorithms like decision tree and random forest classifiers

Univ of Stellenbosch Grant (Febe de Wet) - Voices of Mzansi project aimed to get South Africa's languages launched on the Mozilla Common Voice platform1. To achieve this aim the Common Voice website had to be translated from English into the other official languages and at least 5 000 sentences had to be collected for each language. The collected sentences have to be available in the open domain under CC-0 licensing. Once a languages is launched, the sentences are used as prompts for speech data collection through the Common Voice platform. This unlocks open-source text-to-speech dataset collection and thereby ultimately facilitates access to information in mother tongue, including by illiterate persons and in rural settings.

The Accent Classification Dataset (Ghana) is a collection of audio recordings from native and non-native English speakers across Ghana's diverse regions. Participants read the same three scripts, capturing distinct regional accents and speech patterns. This consistent dataset is valuable for linguistic analysis, accent classification, and speech recognition research focused on Ghanaian English.

The Agrivoltaic system offers a transformative solution for farming communities by providing a means to generate electricity without sacrificing agricultural productivity. The dataset showcases the effectivenesss of Solar PVs and also crop yield under solar PVs and in the open-sun on the same farm.