About Me

I'm a Ph.D. candidate in Electrical Engineering at University of Southern California, working at Signal Analysis and Interpretation Laboratory (SAIL) under Prof. Panayiotis Georgiou and Prof. Shrikanth Narayanan.

My work is focused on understanding and quantification of human emotions and behaviors from speech, language, and vision through deep learning techniques.

My research includes Behavioral Signal Processing, Speech Signal Processing, Affective Computing, Natural Language Processing and Speech Recognition.

Publications

Linking emotions to behaviors through deep transfer learning

Haoqi Li, Brian Baucom, Panayiotis Georgiou PeerJ Computer Science Journal 6:e246, 2020 pdf | bibtex

@article{li2020linking,
  title={Linking emotions to behaviors through deep transfer learning},
  author={Li, Haoqi and Baucom, Brian and Georgiou, Panayiotis},
  journal={PeerJ Computer Science},
  volume={6},
  pages={e246},
  year={2020},
  publisher={PeerJ Inc.}
}

Speaker-invariant Affective Representation Learning via Adversarial Training

Haoqi Li, Ming Tu, Jing Huang, Shrikanth Narayanan, Panayiotis Georgiou IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020 pdf | bibtex

						Not available yet.

Automatic prediction of suicidal risk in military couples using multimodal interaction cues from couples conversations

Sandeep Nallan Chakravarthula, Md Nasir, Shao-Yen Tseng, Haoqi Li, Tae Jin Park, Brian Baucom, Craig J Bryan, Shrikanth Narayanan, Panayiotis Georgiou IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020 pdf | bibtex

						Not available yet.

Predicting Behavior in Cancer-Afflicted Patient and Spouse Interactions using Speech and Language

Haoqi Li*, Sandeep Nallan Chakravarthula*, Shao-Yen Tseng, Maija Reblin, Panayiotis Georgiou Proceedings of Interspeech (Interspeech), 2019 pdf | bibtex

@article{chakravarthulali2019predictingcancer,
title={Predicting Behavior in Cancer-Afflicted Patient and Spouse Interactions using Speech and Language},
author={Chakravarthula, Sandeep Nallan and Li, Haoqi and Tseng, Shao-Yen and Reblin, Maija and Georgiou, Panayiotis},
journal={Proc. Interspeech 2019},
year={2019}
}

"Honey, I Learned to Talk": Multimodal Fusion for Behavior

Haoqi Li*, Shao-Yen Tseng*, Brian Baucom, Panayiotis Georgiou ACM International Conference on Multimodal Interaction (ICMI), 2018 pdf | bibtex

@inproceedings{honeyIlearnedtotalk,
  title={"Honey, I Learned to Talk": Multimodal Fusion for Behavior Analysis},
  author={Tseng, Shao-Yeng and Li, Haoqi and Baucom, Brian and Georgiou, Panayiotis},
  booktitle={Proceedings of the 20th ACM International Conference on Multimodal Interaction},
  year={2018},
  organization={ACM}
}

A Deep Reinforcement Learning Framework for Identifying Funny Scenes in Movies

Haoqi Li, Naveen Kumar, Ruxin Chen, Panayiotis Georgiou IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018 pdf | bibtex

@inproceedings{li2018_a-deep-reinforc,
	Address = {Calgary, Alberta, Canada},
	Author = {Li, Haoqi and Kumar, Naveen and Chen, Ruxin and Georgiou, Panayiotis},
	Booktitle = {Proceedings of IEEE International Conference on Audio, Speech and Signal Processing (ICASSP)},
	Title = {A Deep Reinforcement Learning Framework for Identifying Funny Scenes in Movies},
	Year = {2018}}

Learning from Past Mistakes: Improving Automatic Speech Recognition Output via Noisy-Clean Phrase Context Modeling

Prashanth Gurunath Shivakumar, Haoqi Li, Kevin Knight, and Panayiotis Georgiou APSIPA Transactions on Signal and Information Processing 8 (2019) pdf | bibtex

@article{shivakumar2018_learning-from-p,
	Author = {Shivakumar, Prashanth Gurunath and Li, Haoqi and Knight, Kevin and Georgiou, Panayiotis},
	Eprint = {arXiv:1802.02607},
	Journal = {Eprint arXiv:1802.02607},
	Title = {Learning from Past Mistakes: Improving Automatic Speech Recognition Output via Noisy-Clean Phrase Context Modeling},
	Url = {https://arxiv.org/abs/1802.02607},
	Year = {2018},
	Bdsk-Url-1 = {https://arxiv.org/abs/1802.02607}}

Unsupervised Latent Behavior Manifold Learning from Acoustic Features: audio2behavior

Haoqi Li, Brian Baucom and Panayiotis Georgiou IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2017 pdf | bibtex

@inproceedings{li2017_unsupervised-la,
	Author = {Li, Haoqi and Baucom, Brian and Georgiou, Panayiotis},
	Booktitle = {Proceedings of IEEE International Conference on Audio, Speech and Signal Processing (ICASSP)},
	Month = {March},
	Title = {Unsupervised Latent Behavior Manifold Learning From Acoustic Features: Audio2behavior},
	Year = {2017}}

Sparsely Connected and Disjointly Trained Deep Neural Networks for Low Resource Behavioral Annotation: Acoustic Classification in Couples’ Therapy

Haoqi Li, Brian Baucom and Panayiotis Georgiou Proceedings of Interspeech (Interspeech), 2016 pdf | bibtex

@inproceedings{li2016_sparsely-connec,
	Address = {San Francisco, CA},
	Author = {Li, Haoqi and Baucom, Brian and Georgiou, Panayiotis},
	Booktitle = {Proceedings of Interspeech},
	Month = {September},
	Title = {Sparsely Connected and Disjointly Trained Deep Neural Networks for Low Resource Behavioral Annotation: Acoustic Classification in Couples' Therapy},
	Year = {2016}}

Experience

Internships:
JD AI Research@JD.com, Summer 2018, Mountain View
Acoustic Analysis Research Group @Playstation, SONY, Summer 2017, San Mateo

Teaching Assistant:
USC EE503 Probability for Electrical and Computer Engineers, Spring 2018.
Instructor: Prof. Vijay Kumar
USC EE 660: Machine Learning from Signals: Foundations and Methods, Fall 2019
Instructor: Prof. Keith Jenkins

Reviewer:
ICASSP 2020
IEEE Signal Processing Letters，
APSIPA Transactions on Signal and Information Processing
EURASIP Journal on Audio, Speech, and Music Processing