Applications invited for the following specific domains [ Technology Development Vertical] 1. Machine Translation Hindi and English for Judicial and legal Applications. Details: The Machine Translation System to be based on Deep Neural Networks and should be primarily developed for English-Hindi and Hindi-English. The practice of law involves developing arguments based on the reasoning of courts in previous instances with similar circumstances. This reasoning, known as precedent, is invoked to justify a legal standpoint or discredit the opposition’s arguments. Precedent can be reinforced or dismissed by subsequent court decisions, and is therefore the fundamental point of debate in the application of the law. Consequently, it is imperative for legal practitioners to be well versed in legal precedents relevant to their area of practice, and remain up to date in their knowledge of such. The weight attributed to this information, found mainly through the study of previous cases, makes the ability to access the decisions essential. These processes are largely manual, imposing lengthy production delays on the industry. Furthermore, lawyers dedicate a significant portion of their time and energy to reading and analyzing decisions for the purposes of client representation and developing legal documents. The legal industry is marked by continuous publishing and translation cycles, large volumes of digital content and growing demand to distribute multilingual information. The capacity to administer high volumes of translation requests without delay thereby becomes imperative. India is a multilingual country with great linguistic and cultural diversities. People speak in many different languages, and there is a growing need to make many useful information available in their own vernacular. In a large and highly populated country like India, a huge amount of legal documents are produced daily. The legal domain has continuous publishing cycles, and the growing demand for multilingual information access requires these documents to be translated into a language that the end user understands most. In that respect, the burden of granting access to these information falls on the government which must manage a continuous cycle of document translation needs. A certified translation of a legal judgment takes several months to complete. Necessary & Essential Requirements R&D Team: The research team must consist of - A demonstrated track record of AI, ML, NLP. - The commercial partner to be identified by TIH PET members - The R&D team must support atleast 1 PhD student Project Duration: 2 Years 2. AI and NLP driven Agriculture assistant with Sensors and Embedded Systems Details: The project's objective is to design- (a) an AI driven personal assistant for Farmers to take informed decisions and timely actions. (b) The IoT and sensors must collect the data from the sectors across the farmland, crops, soil, and environment around the farm. (c) This data must be processed via could or other methods to provide alerts to the farmer in multiple languages- initially Hindi and English. (d) The application must also have a voice assistant for the user. Necessary & Essential Requirements R&D Team: The research team must consist of - A demonstrated track record of AI, ML, NLP. - A demonstrated track record Materials Science(s) and delivery of work to industries. Necessary & Essential Requirements Commercial Partner: - A demonstrated track record of Hardware and IoT Development for the industries viz.- Automotive, FPGA cluster test kits, Embedded PCB Design. - The company must be atleast 5 years old in its operations. Project Duration: 2 Years 3. Contactless Sensing for Recognizing Activities of Disabled and Elderly Details: The aging of the world population is becoming increasingly significant. Smart environments have emerged as a prominent research domain, particularly in the realm of providing healthcare and assistive services. These environments utilize diverse sensors, actuators, and computational elements to aid inhabitants in various situations and assist them with daily activities. The objective of this project is to develop a non-contact sensor-based system for recognizing the activities of daily living (ADL) among elderly and disabled individuals during home isolation. The focus is on delivering an unobtrusive solution that eliminates the need for human intervention or wearable sensors. The project objectives are as follows: ● To build low-cost, energy-efficient, and plug-and-play sensor nodes for recognizing the activities of daily living. ● To employ deep learning-based techniques for recognizing the specific ADL being performed by the user and identifies the location where ADL is performed. ● To design low-power wide area networks for energy-efficient communication, ensuring that ADL information is transferred to caregivers promptly. Necessary & Essential Requirements R&D Team: The research team must consist of - ● A demonstrated track record in Smart Sensing, encompassing expertise in sensors, AI/ML, and communication technologies. ● Principal Investigator (PI)/Co-PI should possess industry experience and a proven track record of delivering work to industries. ● The commercial partner will be identified by TIH who has a proven track record in the hardware and software design. Project Duration: 2 Years All applications must be emailed to ceo.tih@iitp.ac.in Last Date For Receiving Applications: 5th December 2023