Shrikant Malviya

• PhD
• Researcher at Indian Institute of Information Technology Allahabad

The rapid advancement of large language models (LLMs) has introduced new challenges in distinguishing human-written text from AI-generated content. In this work, we explored a pipelined approach for AI-generated text detection that includes a feature extraction step (i.e. prompt-based rewriting features inspired by RAIDAR and content-based features...

The aim of this work is to explore the potential of pre-trained vision-language models, e.g. Vision Transformers (ViT), enhanced with advanced data augmentation strategies for the detection of AI-generated images. Our approach leverages a fine-tuned ViT model trained on the Defactify-4.0 dataset, which includes images generated by state-of-the-art...

Background: Technological advances in the smart home have created new opportunities for supporting digital citizens’ well-being and facilitating their empowerment but have enabled new types of complex online harms to develop. Recent statistics have indicated that ‘smart’ technology ownership increases yearly, driven by lower costs and increased acc...

In this paper, we propose a novel method of evaluating text-to-speech systems named “Learning-Based Objective Evaluation” (LBOE), which utilises a set of selected low-level-descriptors (LLD) based features to assess the speech-quality of a TTS model. We have considered Unit selection speech synthesis (USS), Hidden Markov Model speech synthesis (HMM...

In March 29, 2023, the UK Government released a white paper outlining its plans to implement a pro-innovation approach to Artificial Intelligence (AI) regulation and strengthen the UK's position as a global leader in AI. As part of the white paper, the government has developed five key principles to guide regulators. These principles encompass saf...

Dialogue policy is a crucial component in task-oriented Spoken Dialogue Systems (SDSs). As a decision function, it takes the current dialogue state as input and generates appropriate system’s response. In this paper, we explore the reinforcement learning approaches to solve this problem in an Indic language scenario. Recently, Deep Reinforcement Le...

Perceptual computing (Per-C) is a branch of CWW (Computing with words) that assist people in making subjective decisions. Their applications take linguistic inputs (i.e., words) from the user and return a linguistic output (i.e., word). The perception of these linguistic inputs suffers from uncertainties, for which IT2FSs (Interval Type-2 Fuzzy Set...

Impreciseness and uncertainty are the fabrics that make life interesting. For decades, human beings have developed strategies to cope with uncertainties and automate them. In personnel selection for the I.T. field, selectors often find it very difficult to select candidates by going through a set of resumes containing similar kinds of skills. Hence...

Due to the rapid increase in the development of Task-oriented dialogue systems, the need for labelled dialogue corpus has become inevitable. For the Hindi language, there is no such dialogue corpus yet available. As a first attempt, we release a Hindi Dialogue Restaurant Search (HDRS) corpus and compare various state-of-the-art dialogue state track...

Knowledge sharing platforms like Quora have millions and billions of questions. With such a vast number of questions, there will be a lot of duplicates in it. Duplicate questions in these sites are normal, especially with the increasing number of questions asked. These redundant queries reduce efficiency and create repetitive data on the data serve...

Natural Language Generation (NLG) is a crucial component of a Spoken Dialogue System. Its task is to generate utterances with intended attributes like fluency, variation, readability, scalability and adequacy. As the handcrafted models are rigid and tedious to build, people have proposed many statistical and deep-learning based models to bring abou...

Interval Type-2 fuzzy sets (IT2FSs) are used for modeling uncertainty and imprecision in a better way. In a conversation, the information given by humans are mostly words. IT2FSs can be used to provide a suitable mathematical representation of a word. The IT2FSs can be further processed using Computing with the words (CWW) engine to return the IT2F...

In this work, we perform feature selection on fifty-six prosodic features (such as intensity, formants, pause) extracted from the MIT-interview dataset. These features help in rating the various personality traits (Engaged, Excited, Friendly, Speaking Rate) which in turn help to determine an interviewee performance. First, we have demonstrated how...

In this paper, an extended combined approach of phrase based statistical machine translation (SMT), example based MT (EBMT) and rule based MT (RBMT) is proposed to develop a novel hybrid data driven MT system capable of outperforming the baseline SMT, EBMT and RBMT systems from which it is derived. In short, the proposed hybrid MT process is guided...

Automatic speech recognition (ASR) and Text to speech (TTS) are two prominent area of research in human computer interaction nowadays. A set of phonetically rich sentences is in a matter of importance in order to develop these two interactive modules of HCI. Essentially, the set of phonetically rich sentences has to cover all possible phone units d...

In this paper an approach of Semantic Knowledge Extraction (SKE), from a set of research papers, is proposed to develop a system Summarized Research Article Generator (SRAG) which would generate a summarized research article based on the query given by a user. The SRAG stores the semantic knowledge extracted from the query relevant papers in the fo...

Nowadays affective computing is one of the most interesting and challenging research areas among the human computer interaction (HCI) researchers. One of the potential applications of emotion detection is to analyze confidence level of a speaker. In human-computer or human-human interaction systems, speech based confidence level check can provide u...

A lot of work has been done in the field of knowledge extraction from a set of document for multi document summarization (MDS). In this paper an approach of semantic knowledge extraction, from a set of research papers, is used to generate a summarized document. This semantic knowledge is in the form of tree. At each level of the tree score of pa...