Our client is building a tool for agriculture helplines to understand conversations between farmers and agents. This tool will identify crop names, specific issues, and suggested solutions to improve communication and support.

Karya partners with KHPT and USAID, employing 100 individuals to collect diverse speech data for a Kannada voice-based TB knowledge app, benefiting workers through data royalties on the Karya marketplace.

Collecting and validating conversational speech data across multiple domains in 4 languages for IIT Madras

Funded by the Gates Foundation, Karya tackles gender biases in NLU models by collecting 500,000 sentences from rural Indian women, aiming for more inclusive language technologies.

Karya collaborates with GiZ, IIT-Kharagpur, and Microsoft Research to build the first public ASR model for endangered Mundari language, revitalizing Munda cultures.

ICICI Lombard sought to automate car damage detection for insurance claims, partnering with Karya to annotate dented car images with complex polygonal outlines and categorize dents for precise cost estimation using Karya's Labely Platform.

IISc Bangalore collaborated with Karya to enhance camera systems for drone detection in the wilderness, employing workers in Bijapur to annotate camouflaged drone images using Label Studio.

Collecting and validating over 2500 hours of Indic language speech data in 11 languages

Gathered 1 million diverse English newspaper and textbook images for Microsoft India's OCR technology, engaging local libraries and 'kabadiwallas' for efficient data collection.