arXiv subjects

Shrikanth Narayanan

Publications and source records attributed to Shrikanth Narayanan.

At least 145 records · Page 8Linked to original sources

arXiv2023

TrustSER: On the Trustworthiness of Fine-tuning Pre-trained Speech Embeddings For Speech Emotion Recognition

Recent studies have explored the use of pre-trained embeddings for speech emotion recognition (SER), achieving comparable performance to conventional methods that rely on low-level knowledge-inspired acoustic features. These embeddings are often generated from models trained on large-scale speech datasets using self-supervised or weakly-supervised learning objectives. Despite the significant advancements made in SER through the use of pre-trained embeddings, there is a limited understanding of the trustworthiness of these methods, including privacy breaches, unfair performance, vulnerability to adversarial attacks, and computational cost, all of which may hinder the real-world deployment of these systems. In response, we introduce TrustSER, a general framework designed to evaluate the trustworthiness of SER systems using deep learning methods, with a focus on privacy, safety, fairness, and sustainability, offering unique insights into future research in the field of SER. Our code is publicly available under: https://github.com/usc-sail/trust-ser.

cs.SD↗

arXiv2023

Signal Processing Grand Challenge 2023 -- e-Prevention: Sleep Behavior as an Indicator of Relapses in Psychotic Patients

This paper presents the approach and results of USC SAIL's submission to the Signal Processing Grand Challenge 2023 - e-Prevention (Task 2), on detecting relapses in psychotic patients. Relapse prediction has proven to be challenging, primarily due to the heterogeneity of symptoms and responses to treatment between individuals. We address these challenges by investigating the use of sleep behavior features to estimate relapse days as outliers in an unsupervised machine learning setting. We extract informative features from human activity and heart rate data collected in the wild, and evaluate various combinations of feature types and time resolutions. We found that short-time sleep behavior features outperformed their awake counterparts and larger time intervals. Our submission was ranked 3rd in the Task's official leaderboard, demonstrating the potential of such features as an objective and non-invasive predictor of psychotic relapses.

eess.SP↗

arXiv2023

Designing and Evaluating Speech Emotion Recognition Systems: A reality check case study with IEMOCAP

There is an imminent need for guidelines and standard test sets to allow direct and fair comparisons of speech emotion recognition (SER). While resources, such as the Interactive Emotional Dyadic Motion Capture (IEMOCAP) database, have emerged as widely-adopted reference corpora for researchers to develop and test models for SER, published work reveals a wide range of assumptions and variety in its use that challenge reproducibility and generalization. Based on a critical review of the latest advances in SER using IEMOCAP as the use case, our work aims at two contributions: First, using an analysis of the recent literature, including assumptions made and metrics used therein, we provide a set of SER evaluation guidelines. Second, using recent publications with open-sourced implementations, we focus on reproducibility assessment in SER.

cs.SD↗

arXiv2023

Can Knowledge of End-to-End Text-to-Speech Models Improve Neural MIDI-to-Audio Synthesis Systems?

With the similarity between music and speech synthesis from symbolic input and the rapid development of text-to-speech (TTS) techniques, it is worthwhile to explore ways to improve the MIDI-to-audio performance by borrowing from TTS techniques. In this study, we analyze the shortcomings of a TTS-based MIDI-to-audio system and improve it in terms of feature computation, model selection, and training strategy, aiming to synthesize highly natural-sounding audio. Moreover, we conducted an extensive model evaluation through listening tests, pitch measurement, and spectrogram analysis. This work demonstrates not only synthesis of highly natural music but offers a thorough analytical approach and useful outcomes for the community. Our code, pre-trained models, supplementary materials, and audio samples are open sourced at https://github.com/nii-yamagishilab/midi-to-audio.

cs.SD↗

arXiv2023

Contextually-rich human affect perception using multimodal scene information

The process of human affect understanding involves the ability to infer person specific emotional states from various sources including images, speech, and language. Affect perception from images has predominantly focused on expressions extracted from salient face crops. However, emotions perceived by humans rely on multiple contextual cues including social settings, foreground interactions, and ambient visual scenes. In this work, we leverage pretrained vision-language (VLN) models to extract descriptions of foreground context from images. Further, we propose a multimodal context fusion (MCF) module to combine foreground cues with the visual scene and person-based contextual information for emotion prediction. We show the effectiveness of our proposed modular design on two datasets associated with natural scenes and TV shows.

cs.CV↗

arXiv2023

Leveraging Label Correlations in a Multi-label Setting: A Case Study in Emotion

Detecting emotions expressed in text has become critical to a range of fields. In this work, we investigate ways to exploit label correlations in multi-label emotion recognition models to improve emotion detection. First, we develop two modeling approaches to the problem in order to capture word associations of the emotion words themselves, by either including the emotions in the input, or by leveraging Masked Language Modeling (MLM). Second, we integrate pairwise constraints of emotion representations as regularization terms alongside the classification loss of the models. We split these terms into two categories, local and global. The former dynamically change based on the gold labels, while the latter remain static during training. We demonstrate state-of-the-art performance across Spanish, English, and Arabic in SemEval 2018 Task 1 E-c using monolingual BERT-based models. On top of better performance, we also demonstrate improved robustness. Code is available at https://github.com/gchochla/Demux-MEmo.

cs.CL↗

arXiv2023

Using Emotion Embeddings to Transfer Knowledge Between Emotions, Languages, and Annotation Formats

The need for emotional inference from text continues to diversify as more and more disciplines integrate emotions into their theories and applications. These needs include inferring different emotion types, handling multiple languages, and different annotation formats. A shared model between different configurations would enable the sharing of knowledge and a decrease in training costs, and would simplify the process of deploying emotion recognition models in novel environments. In this work, we study how we can build a single model that can transition between these different configurations by leveraging multilingual models and Demux, a transformer-based model whose input includes the emotions of interest, enabling us to dynamically change the emotions predicted by the model. Demux also produces emotion embeddings, and performing operations on them allows us to transition to clusters of emotions by pooling the embeddings of each cluster. We show that Demux can simultaneously transfer knowledge in a zero-shot manner to a new language, to a novel annotation format and to unseen emotions. Code is available at https://github.com/gchochla/Demux-MEmo .

cs.CL↗

arXiv2023

A dataset for Audio-Visual Sound Event Detection in Movies

Audio event detection is a widely studied audio processing task, with applications ranging from self-driving cars to healthcare. In-the-wild datasets such as Audioset have propelled research in this field. However, many efforts typically involve manual annotation and verification, which is expensive to perform at scale. Movies depict various real-life and fictional scenarios which makes them a rich resource for mining a wide-range of audio events. In this work, we present a dataset of audio events called Subtitle-Aligned Movie Sounds (SAM-S). We use publicly-available closed-caption transcripts to automatically mine over 110K audio events from 430 movies. We identify three dimensions to categorize audio events: sound, source, quality, and present the steps involved to produce a final taxonomy of 245 sounds. We discuss the choices involved in generating the taxonomy, and also highlight the human-centered nature of sounds in our dataset. We establish a baseline performance for audio-only sound classification of 34.76% mean average precision and show that incorporating visual information can further improve the performance by about 5%. Data and code are made available for research at https://github.com/usc-sail/mica-subtitle-aligned-movie-sounds

eess.AS↗

arXiv2023

VAuLT: Augmenting the Vision-and-Language Transformer for Sentiment Classification on Social Media

We propose the Vision-and-Augmented-Language Transformer (VAuLT). VAuLT is an extension of the popular Vision-and-Language Transformer (ViLT), and improves performance on vision-and-language (VL) tasks that involve more complex text inputs than image captions while having minimal impact on training and inference efficiency. ViLT, importantly, enables efficient training and inference in VL tasks, achieved by encoding images using a linear projection of patches instead of an object detector. However, it is pretrained on captioning datasets, where the language input is simple, literal, and descriptive, therefore lacking linguistic diversity. So, when working with multimedia data in the wild, such as multimodal social media data, there is a notable shift from captioning language data, as well as diversity of tasks. We indeed find evidence that the language capacity of ViLT is lacking. The key insight and novelty of VAuLT is to propagate the output representations of a large language model (LM) like BERT to the language input of ViLT. We show that joint training of the LM and ViLT can yield relative improvements up to 20% over ViLT and achieve state-of-the-art or comparable performance on VL tasks involving richer language inputs and affective constructs, such as for Target-Oriented Sentiment Classification in TWITTER-2015 and TWITTER-2017, and Sentiment Classification in MVSA-Single and MVSA-Multiple. Our code is available at https://github.com/gchochla/VAuLT.

cs.CV↗

arXiv2022

Exploring Workplace Behaviors through Speaking Patterns using Large-scale Multimodal Wearable Recordings: A Study of Healthcare Providers

Interpersonal spoken communication is central to human interaction and the exchange of information. Such interactive processes involve not only speech and spoken language but also non-verbal cues such as hand gestures, facial expressions, and nonverbal vocalization, that are used to express feelings and provide feedback. These multimodal communication signals carry a variety of information about the people: traits like gender and age as well as about physical and psychological states and behavior. This work uses wearable multimodal sensors to investigate interpersonal communication behaviors focusing on speaking patterns among healthcare providers with a focus on nurses. We analyze longitudinal data collected from $99$ nurses in a large hospital setting over ten weeks. The results indicate that speaking pattern differences across shift schedules and working units. Moreover, results show that speaking patterns combined with physiological measures can be used to predict affect measures and life satisfaction scores. The implementation of this work can be accessed at https://github.com/usc-sail/tiles-audio-arousal.

cs.SD↗

arXiv2022

Audio-Visual Activity Guided Cross-Modal Identity Association for Active Speaker Detection

Active speaker detection in videos addresses associating a source face, visible in the video frames, with the underlying speech in the audio modality. The two primary sources of information to derive such a speech-face relationship are i) visual activity and its interaction with the speech signal and ii) co-occurrences of speakers' identities across modalities in the form of face and speech. The two approaches have their limitations: the audio-visual activity models get confused with other frequently occurring vocal activities, such as laughing and chewing, while the speakers' identity-based methods are limited to videos having enough disambiguating information to establish a speech-face association. Since the two approaches are independent, we investigate their complementary nature in this work. We propose a novel unsupervised framework to guide the speakers' cross-modal identity association with the audio-visual activity for active speaker detection. Through experiments on entertainment media videos from two benchmark datasets, the AVA active speaker (movies) and Visual Person Clustering Dataset (TV shows), we show that a simple late fusion of the two approaches enhances the active speaker detection performance.

cs.MM↗

arXiv2022

A Context-Aware Computational Approach for Measuring Vocal Entrainment in Dyadic Conversations

Vocal entrainment is a social adaptation mechanism in human interaction, knowledge of which can offer useful insights to an individual's cognitive-behavioral characteristics. We propose a context-aware approach for measuring vocal entrainment in dyadic conversations. We use conformers(a combination of convolutional network and transformer) for capturing both short-term and long-term conversational context to model entrainment patterns in interactions across different domains. Specifically we use cross-subject attention layers to learn intra- as well as inter-personal signals from dyadic conversations. We first validate the proposed method based on classification experiments to distinguish between real(consistent) and fake(inconsistent/shuffled) conversations. Experimental results on interactions involving individuals with Autism Spectrum Disorder also show evidence of a statistically-significant association between the introduced entrainment measure and clinical scores relevant to symptoms, including across gender and age groups.

eess.AS↗

arXiv2022

Multimodal Estimation of Change Points of Physiological Arousal in Drivers

Detecting unsafe driving states, such as stress, drowsiness, and fatigue, is an important component of ensuring driving safety and an essential prerequisite for automatic intervention systems in vehicles. These concerning conditions are primarily connected to the driver's low or high arousal levels. In this study, we describe a framework for processing multimodal physiological time-series from wearable sensors during driving and locating points of prominent change in drivers' physiological arousal state. These points of change could potentially indicate events that require just-in-time intervention. We apply time-series segmentation on heart rate and breathing rate measurements and quantify their robustness in capturing change points in electrodermal activity, treated as a reference index for arousal, as well as on self-reported stress ratings, using three public datasets. Our experiments demonstrate that physiological measures are veritable indicators of change points of arousal and perform robustly across an extensive ablation study.

eess.SP↗

arXiv2022

On the Role of Visual Context in Enriching Music Representations

Human perception and experience of music is highly context-dependent. Contextual variability contributes to differences in how we interpret and interact with music, challenging the design of robust models for information retrieval. Incorporating multimodal context from diverse sources provides a promising approach toward modeling this variability. Music presented in media such as movies and music videos provide rich multimodal context that modulates underlying human experiences. However, such context modeling is underexplored, as it requires large amounts of multimodal data along with relevant annotations. Self-supervised learning can help address these challenges by automatically extracting rich, high-level correspondences between different modalities, hence alleviating the need for fine-grained annotations at scale. In this study, we propose VCMR -- Video-Conditioned Music Representations, a contrastive learning framework that learns music representations from audio and the accompanying music videos. The contextual visual information enhances representations of music audio, as evaluated on the downstream task of music tagging. Experimental results show that the proposed framework can contribute additive robustness to audio representations and indicates to what extent musical elements are affected or determined by visual context.

cs.SD↗

arXiv2022

Leveraging Open Data and Task Augmentation to Automated Behavioral Coding of Psychotherapy Conversations in Low-Resource Scenarios

In psychotherapy interactions, the quality of a session is assessed by codifying the communicative behaviors of participants during the conversation through manual observation and annotation. Developing computational approaches for automated behavioral coding can reduce the burden on human coders and facilitate the objective evaluation of the intervention. In the real world, however, implementing such algorithms is associated with data sparsity challenges since privacy concerns lead to limited available in-domain data. In this paper, we leverage a publicly available conversation-based dataset and transfer knowledge to the low-resource behavioral coding task by performing an intermediate language model training via meta-learning. We introduce a task augmentation method to produce a large number of "analogy tasks" - tasks similar to the target one - and demonstrate that the proposed framework predicts target behaviors more accurately than all the other baseline models.

cs.CL↗

arXiv2022

MovieCLIP: Visual Scene Recognition in Movies

Longform media such as movies have complex narrative structures, with events spanning a rich variety of ambient visual scenes. Domain specific challenges associated with visual scenes in movies include transitions, person coverage, and a wide array of real-life and fictional scenarios. Existing visual scene datasets in movies have limited taxonomies and don't consider the visual scene transition within movie clips. In this work, we address the problem of visual scene recognition in movies by first automatically curating a new and extensive movie-centric taxonomy of 179 scene labels derived from movie scripts and auxiliary web-based video datasets. Instead of manual annotations which can be expensive, we use CLIP to weakly label 1.12 million shots from 32K movie clips based on our proposed taxonomy. We provide baseline visual models trained on the weakly labeled dataset called MovieCLIP and evaluate them on an independent dataset verified by human raters. We show that leveraging features from models pretrained on MovieCLIP benefits downstream tasks such as multi-label scene and genre classification of web videos and movie trailers.

cs.CV↗

arXiv2022

Unsupervised active speaker detection in media content using cross-modal information

We present a cross-modal unsupervised framework for active speaker detection in media content such as TV shows and movies. Machine learning advances have enabled impressive performance in identifying individuals from speech and facial images. We leverage speaker identity information from speech and faces, and formulate active speaker detection as a speech-face assignment task such that the active speaker's face and the underlying speech identify the same person (character). We express the speech segments in terms of their associated speaker identity distances, from all other speech segments, to capture a relative identity structure for the video. Then we assign an active speaker's face to each speech segment from the concurrently appearing faces such that the obtained set of active speaker faces displays a similar relative identity structure. Furthermore, we propose a simple and effective approach to address speech segments where speakers are present off-screen. We evaluate the proposed system on three benchmark datasets -- Visual Person Clustering dataset, AVA-active speaker dataset, and Columbia dataset -- consisting of videos from entertainment and broadcast media, and show competitive performance to state-of-the-art fully supervised methods.

eess.IV↗

arXiv2022

The 2022 Far-field Speaker Verification Challenge: Exploring domain mismatch and semi-supervised learning under the far-field scenario

FFSVC2022 is the second challenge of far-field speaker verification. FFSVC2022 provides the fully-supervised far-field speaker verification to further explore the far-field scenario and proposes semi-supervised far-field speaker verification. In contrast to FFSVC2020, FFSVC2022 focus on the single-channel scenario. In addition, a supplementary set for the FFSVC2020 dataset is released this year. The supplementary set consists of more recording devices and has the same data distribution as the FFSVC2022 evaluation set. This paper summarizes the FFSVC 2022, including tasks description, trial designing details, a baseline system and a summary of challenge results. The challenge results indicate substantial progress made in the field but also present that there are still difficulties with the far-field scenario.

eess.AS↗