Sarala Padi

Research efforts include building and evaluating cutting-edge deep learning and machine learning models for multimodal emotion recognition, medical imaging, and drug design analyses. Additionally, my work encompasses the advancement of graph and natural language processing (NLP) methods for drug design to generate amino acid sequences representing potentially inhibitory peptides efficiently.

Awards

In January-May 2013, I was awarded the Outstanding Teaching Assistant award for the course Principles of Communications by the Department of Computer Science at IIT Madras, India. I have been a member of SigmaXi since 2016 and a member of ACM since 2020. Additionally, I have been serving as a reviewer for ACM Multimedia since 2020, ACM Multimedia Asia since 2020, ICASSP since 2020, and Interspeech since 2018.

Selected Publications

Quantitative Bright-Field Microscopy Combined with Deep Neural Networks Predict Live Tissue Function

February 29, 2020

Author(s)

Carl Simon Jr., Nicholas J. Schaub, Petru S. Manescu, Sarala Padi, Mylene Simon, Peter Bajcsy, Nathan A. Hotaling, Joe Chalfoun, Mohamed Ouladi, Qin Wan, Kapil Bharti, Ruchi Sharma

Progressive increases in the number of cell therapies in the preclinical and clinical phases has prompted the need for reliable and non-invasive assays to

Cell Image Segmentation using Generative Adversarial Networks, Transfer Learning, and Augmentations

June 17, 2019

Author(s)

Michael P. Majurski, Petru S. Manescu, Sarala Padi, Nicholas J. Schaub, Nathan A. Hotaling, Carl Simon Jr., Peter Bajcsy

We address the problem of segmenting cell contours from microscopy images of human induced pluripotent Retinal Pigment Epithelial stem cells (iRPE) using

Multimodal Emotion Recognition using Transfer Learning from Speaker Recognition and BERT based models

July 1, 2022

Author(s)

Sarala Padi, Omid Sadjadi, Dinesh Manocha, Ram Sriram

Automatic emotion recognition plays a key role in computer-human interaction as it has the potential to enrich the next generation artificial intelligence with

Improved Speech Emotion Recognition using Transfer Learning and Spectrogram Augmentation

August 5, 2021

Author(s)

Sarala Padi, Omid Sadjadi, Ram D. Sriram, Dinesh Manocha

Automatic speech emotion recognition (SER) is a challenging task that plays a crucial role in natural human-computer interaction. One of the main challenges in

Augmenting Deep Learning Models for Speech Emotion Recognition

October 19, 2020

Author(s)

Ram Sriram, Dinesh Manocha, Sarala Padi

We present a Multi-Window Data Augmentation (MWA-SER) approach for speech emotion recognition. MWA-SER is a unimodal approach that focuses on two key concepts

Publications

Meta-model for ADMET Property Prediction Analysis

December 7, 2023

Author(s)

Sarala Padi, Antonio Cardone, Ram Sriram

In drug discovery analysis ADMET properties, such as chemical absorption, distribution, metabolism, excretion, and toxicity, play a critical role. They allow

Multimodal Emotion Recognition using Transfer Learning from Speaker Recognition and BERT based models

July 1, 2022

Author(s)

Sarala Padi, Omid Sadjadi, Dinesh Manocha, Ram Sriram

Automatic emotion recognition plays a key role in computer-human interaction as it has the potential to enrich the next generation artificial intelligence with

Improved Speech Emotion Recognition using Transfer Learning and Spectrogram Augmentation

August 5, 2021

Author(s)

Sarala Padi, Omid Sadjadi, Ram D. Sriram, Dinesh Manocha

Automatic speech emotion recognition (SER) is a challenging task that plays a crucial role in natural human-computer interaction. One of the main challenges in

Deconstructing UML, Part 1: Modeling Classes with Categories

May 13, 2021

Author(s)

Sarala Padi, Spencer J. Breiner, Eswaran Subrahmanian, Ram D. Sriram

In this paper we analyze elements of the Unified Modeling Language (UML), specifically the class diagram, and propose an simplified alternative language based

Comparing Footwear Impressions that are Close Non-Matches Using Correlation-Based Approaches

May 1, 2021

Author(s)

Gautham Venkatasubramanian, Vighnesh Hegde, Sarala Padi, Hariharan K. Iyer, Martin Herman

Forensic activities related to footwear evidence may be broadly classified into the following two categories: (1) intelligence gathering and (2) evidential

View more Publications

Sarala Padi (IntlAssoc)

Computer Scientist

Awards

Selected Publications

Quantitative Bright-Field Microscopy Combined with Deep Neural Networks Predict Live Tissue Function

Cell Image Segmentation using Generative Adversarial Networks, Transfer Learning, and Augmentations

Multimodal Emotion Recognition using Transfer Learning from Speaker Recognition and BERT based models

Improved Speech Emotion Recognition using Transfer Learning and Spectrogram Augmentation

Augmenting Deep Learning Models for Speech Emotion Recognition

Publications

Meta-model for ADMET Property Prediction Analysis

Multimodal Emotion Recognition using Transfer Learning from Speaker Recognition and BERT based models

Improved Speech Emotion Recognition using Transfer Learning and Spectrogram Augmentation

Deconstructing UML, Part 1: Modeling Classes with Categories

Comparing Footwear Impressions that are Close Non-Matches Using Correlation-Based Approaches