Picture for Shrikanth Narayanan

Shrikanth Narayanan

User-Level Differential Privacy against Attribute Inference Attack of Speech Emotion Recognition in Federated Learning

Apr 05, 2022
Tiantian Feng, Raghuveer Peri, Shrikanth Narayanan

Figure 1 for User-Level Differential Privacy against Attribute Inference Attack of Speech Emotion Recognition in Federated Learning

Figure 2 for User-Level Differential Privacy against Attribute Inference Attack of Speech Emotion Recognition in Federated Learning

Figure 3 for User-Level Differential Privacy against Attribute Inference Attack of Speech Emotion Recognition in Federated Learning

Access Paper or Ask Questions

Multimodal Clustering with Role Induced Constraints for Speaker Diarization

Apr 01, 2022
Nikolaos Flemotomos, Shrikanth Narayanan

Figure 1 for Multimodal Clustering with Role Induced Constraints for Speaker Diarization

Figure 2 for Multimodal Clustering with Role Induced Constraints for Speaker Diarization

Figure 3 for Multimodal Clustering with Role Induced Constraints for Speaker Diarization

Figure 4 for Multimodal Clustering with Role Induced Constraints for Speaker Diarization

* Submitted at Interspeech 2022

Access Paper or Ask Questions

Using Active Speaker Faces for Diarization in TV shows

Mar 30, 2022
Rahul Sharma, Shrikanth Narayanan

Figure 1 for Using Active Speaker Faces for Diarization in TV shows

Figure 2 for Using Active Speaker Faces for Diarization in TV shows

Figure 3 for Using Active Speaker Faces for Diarization in TV shows

Figure 4 for Using Active Speaker Faces for Diarization in TV shows

* Submitted to Interspeech 2022

Access Paper or Ask Questions

Mel Frequency Spectral Domain Defenses against Adversarial Attacks on Speech Recognition Systems

Mar 29, 2022
Nicholas Mehlman, Anirudh Sreeram, Raghuveer Peri, Shrikanth Narayanan

Figure 1 for Mel Frequency Spectral Domain Defenses against Adversarial Attacks on Speech Recognition Systems

Figure 2 for Mel Frequency Spectral Domain Defenses against Adversarial Attacks on Speech Recognition Systems

Figure 3 for Mel Frequency Spectral Domain Defenses against Adversarial Attacks on Speech Recognition Systems

Figure 4 for Mel Frequency Spectral Domain Defenses against Adversarial Attacks on Speech Recognition Systems

* This paper is 5 pages long and was submitted to Interspeech 2022

Access Paper or Ask Questions

Audio visual character profiles for detecting background characters in entertainment media

Mar 21, 2022
Rahul Sharma, Shrikanth Narayanan

Figure 1 for Audio visual character profiles for detecting background characters in entertainment media

Figure 2 for Audio visual character profiles for detecting background characters in entertainment media

Figure 3 for Audio visual character profiles for detecting background characters in entertainment media

Figure 4 for Audio visual character profiles for detecting background characters in entertainment media

* submitted to ICIP 2022

Access Paper or Ask Questions

To train or not to train adversarially: A study of bias mitigation strategies for speaker recognition

Mar 17, 2022
Raghuveer Peri, Krishna Somandepalli, Shrikanth Narayanan

Figure 1 for To train or not to train adversarially: A study of bias mitigation strategies for speaker recognition

Figure 2 for To train or not to train adversarially: A study of bias mitigation strategies for speaker recognition

Figure 3 for To train or not to train adversarially: A study of bias mitigation strategies for speaker recognition

Figure 4 for To train or not to train adversarially: A study of bias mitigation strategies for speaker recognition

* Preprint submitted to Computer Speech and Language (Elsevier)

Access Paper or Ask Questions

Semi-FedSER: Semi-supervised Learning for Speech Emotion Recognition On Federated Learning using Multiview Pseudo-Labeling

Mar 15, 2022
Tiantian Feng, Shrikanth Narayanan

Figure 1 for Semi-FedSER: Semi-supervised Learning for Speech Emotion Recognition On Federated Learning using Multiview Pseudo-Labeling

Figure 2 for Semi-FedSER: Semi-supervised Learning for Speech Emotion Recognition On Federated Learning using Multiview Pseudo-Labeling

Figure 3 for Semi-FedSER: Semi-supervised Learning for Speech Emotion Recognition On Federated Learning using Multiview Pseudo-Labeling

* This paper was submitted to Insterspeech 2022 for review

Access Paper or Ask Questions

Understanding of Emotion Perception from Art

Oct 13, 2021
Digbalay Bose, Krishna Somandepalli, Souvik Kundu, Rimita Lahiri, Jonathan Gratch, Shrikanth Narayanan

Figure 1 for Understanding of Emotion Perception from Art

Figure 2 for Understanding of Emotion Perception from Art

Figure 3 for Understanding of Emotion Perception from Art

Figure 4 for Understanding of Emotion Perception from Art

* 5 pages, 5 figures. Accepted at ICCV2021: 4th Workshop on Closing the loop between Vision and Language

Access Paper or Ask Questions

Cross Domain Emotion Recognition using Few Shot Knowledge Transfer

Oct 11, 2021
Justin Olah, Sabyasachee Baruah, Digbalay Bose, Shrikanth Narayanan

Figure 1 for Cross Domain Emotion Recognition using Few Shot Knowledge Transfer

Figure 2 for Cross Domain Emotion Recognition using Few Shot Knowledge Transfer

Figure 3 for Cross Domain Emotion Recognition using Few Shot Knowledge Transfer

Figure 4 for Cross Domain Emotion Recognition using Few Shot Knowledge Transfer

* 5 pages, 4 figures

Access Paper or Ask Questions