Opensource AI
VaaniConnect
VaaniConnect is a multilingual conversational AI platform bridging communication gaps in South Indian languages (Telugu, Tamil, Kannada) for rural communities, Offering voice-to-text, text-to-speech, and real-time translation, it supports local governance, education, and healthcare by addressing regional language barriers in collaboration with Purdue University
- Collaboration with Purdue University
- Collaborative Learning and knowledge exchange
- Working on Real-Time Problem which benefits local and global communities
- Mentored by Dr Kameswara Rao
Open Data Portal
From enhancing geographic accessibility to empowering community-driven development, OpenStreetMap (OSM) data can benefit citizens, organizations – and governments themselves.
- Contributing to street level imagery across india focused on AP
- Dense annotations of indian cityscape
- Rural + urban diverse cityscapes
- Building open and free models for indian cityscape segmentation
Localization
Adaptation of content, tools, and platforms to make them accessible, relevant, and usable for diverse communities around the world. It plays a critical role in democratizing knowledge and ensuring inclusivity
- Improving Localization in Digital Content
- Developing Region-Specific Solutions
- Supporting Localization Efforts through Open Platforms
- Building Localized Open Data Repositories
- Contributed to Bigbluebutton, Plinth, KDE etc Projects
Swecha Voice
Developing the First Open Source AI for the Telugu Language that can communicate in telugu and dialects.
We focused on building datasets by gathering voice samples from individuals of various ages, dialects, and genders which will be available in open source, allowing researchers and developers to access it for free.
- Building a dataset for the people, by the people
- Community based Model
- Localization
