Evaluating Multilingual LLMs at Scale
For Microsoft Research, Karya completed one of the largest multilingual human evaluations of LLMs within three weeks.
Building the Largest Gender-Intentional AI Corpora
Karya, in collaboration with a major philanthropic donor is working to build the largest gender-intentional corpora in Indic languages yet, while providing livelihood opportunities for rural women from low-income communities.
24K+
hours of Voice Data
6
Languages
30K+
Unique Female Speakers
100
Districts across India
Under this project, Karya will onboard and work with 30,000 participants to generate gender-bias-specific speech data in 6 languages spoken by over 900M people & covering over 75% of India.
The end goal is to build a Gender-Intentional Data Tool that can flag biased sentences in AI corpora across the 6 languages, while building the largest gender-intentional AI corpora in Indic language history.
Karya’s workforce can rebuild your training datasets to improve inclusion, with speed and at scale.
Related
Conversational Data and Call Center Data Collection(2400 hours)
Collecting and validating conversational speech data across multiple domains in 11 languages for Microsoft