Tiantian Feng

This amazing picture was taken by my wife at Skyline Trail, Mount Rainier, 2023.

PS: Prepare enough water for this trail but make sure you are not the one carrying it.

(I am on a unique parental leaving situation, and I would be having limited bandwidth to respond your emails till October/November. Sorry that if I cannot respond your email in a timely manner.)

My name is Tiantian Feng, and I grew up in Leshan, China, which is famously for Leshan Giant Buddha. I recently completed my Ph.D. in the Thomas Lord Department of Computer Science at University of Southern California in 2024, and I am currently a postdoc researcher in SAIL lab at USC. I am fortunate to be advised by Professor Shrikanth Narayanan, a globally recognized scientist in speech modeling, linguistics, affective computing, and human understanding.

My research focuses on leveraging sensors and computational methods for understanding natural human behaviors in healthcare applications across life span. Notably, I have involved in projects studying children with autism, aging population with cognitive decline, mental health (e.g. Postpartum depression (PPD)), people with hearing loss, and workplace behavior of hospital professionals. My vision is to develop full-stack AI technology that explores sensings, learning, understanding, and assisting human in a wide range of healthcare-centered applications.

I also have a particular interest in building technology that is private and broadly accessible. I have been focused on developing datasets and benchmark that are shareable across different community. My research invovles applications such as speech understanding, wearable computing, multimodal understanding, and bio-signal processing, etc. Additionally, I have hands-on experience in industrial sensor design and deploying sensors in research studies. I have interned at both Meta and Amazon as the research scientist.

I am actively looking for tenure-track or similar positions in academia. Don’t hesitate to contact me if you would like to collaborate.

News

Jan 20, 2026	Four of my co-authored papers have been accepted to ICASSP! The topics include speech privacy, speech synthesis, speech encoding, and applications of speech emotion recognition.
Jan 10, 2026	My preprint paper VoxCog: Towards End-to-End Multilingual Cognitive Impairment Classification through Dialectal Knowledge is now available online.
Dec 31, 2025	My co-authored paper Speech acoustics to rt-MRI articulatory dynamics inversion with video diffusion model. has been published at Computer Speech & Language (2025).
Dec 10, 2025	I will give a talk on-Developing Robust Speaker Diarization for Child-Adult Dyadic Interaction in ASRU 2025 Satellite Workshop-AI for Children’s Speech and Language!
Nov 23, 2025	Our Voxlect benchmark has been accepted to 2026 KDD Dataset and Benchmark track, congratulations to all co-authors!
Nov 14, 2025	I will be giving an invited talk on “Toward Human-Centered Computing for Behavioral Understanding in Healthcare Applications” to THE Ohio State University Talk Details!
Aug 20, 2025	7 papers accepted and will be presented at INTERSPEECH 2025! The topics include speech datasets for science, child speech classification, child speech-LLMs, child speech ASR, multimodal child behavior understanding, categorized speech emotion recognition, and arousal-valence speech recognition.
Aug 20, 2025	Our work has won the 2nd place in 2025 INTERSPEECH Speech Emotion Recognition Challenge - Task1!
Aug 20, 2025	Our work has won the 1st place in 2025 INTERSPEECH Speech Emotion Recognition Challenge - Task2!
Mar 20, 2025	4 papers accepted and will be presented at ICASSP 2025!

Selected Publications

KDD Benchmark

Voxlect: A Speech Foundation Model Benchmark for Modeling Dialects and Regional Languages Around the Globe

Tiantian Feng , Kevin Huang , Anfeng Xu , Xuan Shi , Thanathai Lertpetchpun , Jihwan Lee , Yoonjeong Lee , Dani Byrd , and Shrikanth Narayanan

Accepted to 2026 KDD Dataset and Benchmark Track, 2026

Bib

@article{feng2025voxlect,
  title = {Voxlect: A Speech Foundation Model Benchmark for Modeling Dialects and Regional Languages Around the Globe},
  author = {Feng, Tiantian and Huang, Kevin and Xu, Anfeng and Shi, Xuan and Lertpetchpun, Thanathai and Lee, Jihwan and Lee, Yoonjeong and Byrd, Dani and Narayanan, Shrikanth},
  journal = {Accepted to 2026 KDD Dataset and Benchmark Track},
  year = {2026},
  abbr2 = {Benchmark},
}

KDD 2023 Benchmark

FedMultimodal: A Benchmark for Multimodal Federated Learning

Tiantian Feng , Digbalay Bose , Tuo Zhang , Rajat Hebbar , Anil Ramakrishna , Rahul Gupta , Mi Zhang , Salman Avestimehr , and Shrikanth Narayanan

In 2023 KDD , 2023

Bib Code

@inproceedings{feng2023fedmultimodal,
  title = {FedMultimodal: A Benchmark for Multimodal Federated Learning},
  author = {Feng, Tiantian and Bose, Digbalay and Zhang, Tuo and Hebbar, Rajat and Ramakrishna, Anil and Gupta, Rahul and Zhang, Mi and Avestimehr, Salman and Narayanan, Shrikanth},
  booktitle = {2023 KDD},
  pages = {4035--4045},
  year = {2023},
  abbr2 = {Benchmark},
}

Preprint Digital Health

VoxCog: Towards End-to-End Multilingual Cognitive Impairment Classification through Dialectal Knowledge

Tiantian Feng , Anfeng Xu , Jinkook Lee , and Shrikanth Narayanan

Under review, 2026

Bib

@article{feng2026voxcog,
  title = {VoxCog: Towards End-to-End Multilingual Cognitive Impairment Classification through Dialectal Knowledge},
  author = {Feng, Tiantian and Xu, Anfeng and Lee, Jinkook and Narayanan, Shrikanth},
  journal = {Under review},
  year = {2026},
  abbr2 = {Digital Health},
}

Preprint Benchmark

Vox-Profile: A Speech Foundation Model Benchmark for Characterizing Diverse Speaker and Speech Traits

Tiantian Feng , Jihwan Lee , Anfeng Xu , Yoonjeong Lee , Thanathai Lertpetchpun , Xuan Shi , Helin Wang , Thomas Thebaud , Laureano Moro-Velazquez , Dani Byrd , and 1 more author

Under review, 2025

Bib

@article{feng2025vox,
  title = {Vox-Profile: A Speech Foundation Model Benchmark for Characterizing Diverse Speaker and Speech Traits},
  author = {Feng, Tiantian and Lee, Jihwan and Xu, Anfeng and Lee, Yoonjeong and Lertpetchpun, Thanathai and Shi, Xuan and Wang, Helin and Thebaud, Thomas and Moro-Velazquez, Laureano and Byrd, Dani and others},
  journal = {Under review},
  year = {2025},
  abbr2 = {Benchmark},
}

Preprint Dataset

TILES-2018 Sleep Benchmark Dataset: A Longitudinal Wearable Sleep Data Set of Hospital Workers for Modeling and Understanding Sleep Behaviors

Tiantian Feng , Brandon M Booth , Karel Mundnich , Emily Zhou , Benjamin Girault , Kristina Lerman , and Shrikanth Narayanan

Under Review, 2025

Bib

@article{feng2025tiles,
  title = {TILES-2018 Sleep Benchmark Dataset: A Longitudinal Wearable Sleep Data Set of Hospital Workers for Modeling and Understanding Sleep Behaviors},
  author = {Feng, Tiantian and Booth, Brandon M and Mundnich, Karel and Zhou, Emily and Girault, Benjamin and Lerman, Kristina and Narayanan, Shrikanth},
  journal = {Under Review},
  year = {2025},
  abbr2 = {Dataset},
  abbr3 = {Digital Health},
}

Scientific Data Dataset

TILES-2018, a longitudinal physiologic and behavioral data set of hospital workers

Karel Mundnich , Brandon M Booth , Michelle l’Hommedieu , Tiantian Feng , Benjamin Girault , Justin L’hommedieu , Mackenzie Wildman , Sophia Skaaden , Amrutha Nadarajan , Jennifer L Villatte , and 1 more author

Scientific Data, 2020

Bib Website

@article{mundnich2020tiles,
  title = {TILES-2018, a longitudinal physiologic and behavioral data set of hospital workers},
  author = {Mundnich, Karel and Booth, Brandon M and l'Hommedieu, Michelle and Feng, Tiantian and Girault, Benjamin and L'hommedieu, Justin and Wildman, Mackenzie and Skaaden, Sophia and Nadarajan, Amrutha and Villatte, Jennifer L and others},
  journal = {Scientific Data},
  volume = {7},
  number = {1},
  pages = {354},
  year = {2020},
  publisher = {Nature Publishing Group UK London},
  abbr2 = {Dataset},
  abbr3 = {Digital Health},
}

Scientific Data Dataset

TILES-2019: A longitudinal physiologic and behavioral data set of medical residents in an intensive care unit

Joanna C Yau , Benjamin Girault , Tiantian Feng , Karel Mundnich , Amrutha Nadarajan , Brandon M Booth , Emilio Ferrara , Kristina Lerman , Eric Hsieh , and Shrikanth Narayanan

Scientific Data, 2022

Bib Website

@article{yau2022tiles,
  title = {TILES-2019: A longitudinal physiologic and behavioral data set of medical residents in an intensive care unit},
  author = {Yau, Joanna C and Girault, Benjamin and Feng, Tiantian and Mundnich, Karel and Nadarajan, Amrutha and Booth, Brandon M and Ferrara, Emilio and Lerman, Kristina and Hsieh, Eric and Narayanan, Shrikanth},
  journal = {Scientific Data},
  volume = {9},
  number = {1},
  pages = {536},
  year = {2022},
  publisher = {Nature Publishing Group UK London},
  abbr2 = {Dataset},
  abbr3 = {Digital Health},
}

Challenge Winner Speech Emotion

Developing a Top-tier Framework in Naturalistic Conditions Challenge for Categorized Emotion Prediction: From Speech Foundation Models and Learning Objective to Data Augmentation and Engineering Choices

Tiantian Feng , Thanathai Lertpetchpun , Dani Byrd , and Shrikanth Narayanan

In Interspeech 2025 , 2025

Bib

@inproceedings{feng25_interspeech,
  title = {{Developing a Top-tier Framework in Naturalistic Conditions Challenge for Categorized Emotion Prediction: From Speech Foundation Models and Learning Objective to Data Augmentation and Engineering Choices}},
  author = {Feng, Tiantian and Lertpetchpun, Thanathai and Byrd, Dani and Narayanan, Shrikanth},
  year = {2025},
  booktitle = {{Interspeech 2025}},
  pages = {4643--4647},
  doi = {10.21437/Interspeech.2025-1163},
  issn = {2958-1796},
  abbr2 = {Speech Emotion},
}

Digital Health INTERSPEECH 2025

Egocentric Speaker Classification in Child-Adult Dyadic Interactions: From Sensing to Computational Modeling

Tiantian Feng , Anfeng Xu , Xuan Shi , Somer Bishop , and Shrikanth Narayanan

In Interspeech 2025 , 2025

Bib

@inproceedings{feng25b_interspeech,
  title = {{Egocentric Speaker Classification in Child-Adult Dyadic Interactions: From Sensing to Computational Modeling}},
  author = {Feng, Tiantian and Xu, Anfeng and Shi, Xuan and Bishop, Somer and Narayanan, Shrikanth},
  year = {2025},
  booktitle = {{Interspeech 2025}},
  pages = {2835--2839},
  doi = {10.21437/Interspeech.2025-1216},
  issn = {2958-1796},
  abbr2 = {INTERSPEECH 2025},
}

ACM Multimedia Dataset

MM-AU: Towards Multimodal Understanding of Advertisement Videos

Digbalay Bose , Rajat Hebbar , Tiantian Feng , Krishna Somandepalli , Anfeng Xu , and Shrikanth Narayanan

In Proceedings of the 31st ACM International Conference on Multimedia , 2023

Bib

@inproceedings{bose2023mm,
  title = {MM-AU: Towards Multimodal Understanding of Advertisement Videos},
  author = {Bose, Digbalay and Hebbar, Rajat and Feng, Tiantian and Somandepalli, Krishna and Xu, Anfeng and Narayanan, Shrikanth},
  booktitle = {Proceedings of the 31st ACM International Conference on Multimedia},
  pages = {86--95},
  year = {2023},
  abbr2 = {Dataset},
}

Preprint Generative AI

Can Text-to-image Model Assist Multi-modal Learning for Visual Recognition with Visual Modality Missing?

Tiantian Feng , Daniel Yang , Digbalay Bose , and Shrikanth Narayanan

arXiv preprint arXiv:2402.09036, 2024

Bib

@article{feng2024can,
  title = {Can Text-to-image Model Assist Multi-modal Learning for Visual Recognition with Visual Modality Missing?},
  author = {Feng, Tiantian and Yang, Daniel and Bose, Digbalay and Narayanan, Shrikanth},
  journal = {arXiv preprint arXiv:2402.09036},
  year = {2024},
  abbr2 = {Generative AI},
}

Voice Privacy Generative AI

Unlocking Foundation Models for Privacy-Enhancing Speech Understanding: An Early Study on Low Resource Speech Training Leveraging Label-guided Synthetic Speech Content

Tiantian Feng , Digbalay Bose , Xuan Shi , and Shrikanth Narayanan

arXiv preprint arXiv:2306.07791, 2023

Bib

@article{feng2023unlocking,
  title = {Unlocking Foundation Models for Privacy-Enhancing Speech Understanding: An Early Study on Low Resource Speech Training Leveraging Label-guided Synthetic Speech Content},
  author = {Feng, Tiantian and Bose, Digbalay and Shi, Xuan and Narayanan, Shrikanth},
  journal = {arXiv preprint arXiv:2306.07791},
  year = {2023},
  abbr2 = {Generative AI},
}

CVPRW Generative AI

GPT-FL: Generative pre-trained model-assisted federated learning

Tuo Zhang* , Tiantian Feng* , Samiul Alam , Mi Zhang , Shrikanth S Narayanan , and Salman Avestimehr

arXiv preprint arXiv:2306.02210, 2023

Bib

@article{zhang2023gpt,
  title = {GPT-FL: Generative pre-trained model-assisted federated learning},
  author = {Zhang*, Tuo and Feng*, Tiantian and Alam, Samiul and Zhang, Mi and Narayanan, Shrikanth S and Avestimehr, Salman},
  journal = {arXiv preprint arXiv:2306.02210},
  year = {2023},
  abbr2 = {Generative AI},
}

Journal Trustworthiness

A Review of Speech-centric Trustworthy Machine Learning: Privacy, Safety, and Fairness

Tiantian Feng , Rajat Hebbar , Nicholas Mehlman , Xuan Shi , Aditya Kommineni , and Shrikanth Narayanan

APSIPA Transactions on Signal and Information Processing, 2023

Bib

@article{feng2023review,
  title = {A Review of Speech-centric Trustworthy Machine Learning: Privacy, Safety, and Fairness},
  author = {Feng, Tiantian and Hebbar, Rajat and Mehlman, Nicholas and Shi, Xuan and Kommineni, Aditya and Narayanan, Shrikanth},
  journal = {APSIPA Transactions on Signal and Information Processing},
  volume = {12},
  number = {3},
  year = {2023},
  publisher = {Now Publishers, Inc.},
  abbr2 = {Trustworthiness},
}

ICASSP 2023 Federated Learning

FedAudio: A federated learning benchmark for audio tasks

Tuo Zhang , Tiantian Feng , Samiul Alam , Sunwoo Lee , Mi Zhang , Shrikanth S Narayanan , and Salman Avestimehr

In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , 2023

Bib Code

@inproceedings{zhang2023fedaudio,
  title = {FedAudio: A federated learning benchmark for audio tasks},
  author = {Zhang, Tuo and Feng, Tiantian and Alam, Samiul and Lee, Sunwoo and Zhang, Mi and Narayanan, Shrikanth S and Avestimehr, Salman},
  booktitle = {ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  pages = {1--5},
  year = {2023},
  organization = {IEEE},
  abbr2 = {Federated Learning},
}

INTERSPEECH 2022 Federated Learning

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

Tiantian Feng , Raghuveer Peri , and Shrikanth Narayanan

In Proc. Interspeech 2022 , 2022

Bib Code

@inproceedings{feng22b_interspeech,
  author = {Feng, Tiantian and Peri, Raghuveer and Narayanan, Shrikanth},
  title = {{User-Level Differential Privacy against Attribute Inference Attack of Speech Emotion Recognition on Federated Learning}},
  year = {2022},
  booktitle = {Proc. Interspeech 2022},
  pages = {5055--5059},
  doi = {10.21437/Interspeech.2022-10060},
  abbr2 = {Federated Learning},
}

ACII 2023 Foundation Model

PEFT-SER: On the use of parameter efficient transfer learning approaches for speech emotion recognition using pre-trained speech models

Tiantian Feng , and Shrikanth Narayanan

In 2023 11th International Conference on Affective Computing and Intelligent Interaction (ACII) , 2023

Bib Code

@inproceedings{feng2023peft,
  title = {PEFT-SER: On the use of parameter efficient transfer learning approaches for speech emotion recognition using pre-trained speech models},
  author = {Feng, Tiantian and Narayanan, Shrikanth},
  booktitle = {2023 11th International Conference on Affective Computing and Intelligent Interaction (ACII)},
  pages = {1--8},
  year = {2023},
  organization = {IEEE},
  abbr2 = {Foundation Model},
}

Scientific Reports Wearables

A multimodal analysis of physical activity, sleep, and work shift in nurses with wearable sensor data

Tiantian Feng , Brandon M Booth , Brooke Baldwin-Rodrı́guez , Felipe Osorno , and Shrikanth Narayanan

Scientific reports, 2021

Bib Code

@article{feng2021multimodal,
  title = {A multimodal analysis of physical activity, sleep, and work shift in nurses with wearable sensor data},
  author = {Feng, Tiantian and Booth, Brandon M and Baldwin-Rodr{\'\i}guez, Brooke and Osorno, Felipe and Narayanan, Shrikanth},
  journal = {Scientific reports},
  volume = {11},
  number = {1},
  pages = {8693},
  year = {2021},
  publisher = {Nature Publishing Group UK London},
  abbr2 = {Wearables},
}

Mobisys Workshop Wearables

Tiles Audio Recorder: An Unobtrusive Wearable Solution to Track Audio Activity

Tiantian Feng , Amrutha Nadarajan , Colin Vaz , Brandon Booth , and Shrikanth Narayanan

In Proceedings of the 4th ACM Workshop on Wearable Systems and Applications , 2018

Bib Code

@inproceedings{feng2018tiles,
  title = {Tiles Audio Recorder: An Unobtrusive Wearable Solution to Track Audio Activity},
  author = {Feng, Tiantian and Nadarajan, Amrutha and Vaz, Colin and Booth, Brandon and Narayanan, Shrikanth},
  booktitle = {Proceedings of the 4th ACM Workshop on Wearable Systems and Applications},
  pages = {33--38},
  year = {2018},
  abbr2 = {Wearables},
  abbr3 = {Digital Health},
}

ICASSP 2019 Wearables

Discovering optimal variable-length time series motifs in large-scale wearable recordings of human bio-behavioral signals

Tiantian Feng , and Shrikanth S Narayanan

In ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , 2019

Bib

@inproceedings{feng2019discovering,
  title = {Discovering optimal variable-length time series motifs in large-scale wearable recordings of human bio-behavioral signals},
  author = {Feng, Tiantian and Narayanan, Shrikanth S},
  booktitle = {ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  pages = {7615--7619},
  year = {2019},
  organization = {IEEE},
  abbr2 = {Wearables}
}