Anfeng Xu (Alex)

RTH 318

University of Southern California

Los Angeles, CA

Hello! I am a Ph.D. candidate in Electrical and Computer Engineering at University of Southern California, supervised by Shrikanth (Shri) Narayanan. My research interests broadly span speech and language technologies, including speech language models, automatic speech recognition (ASR), speaker diarization, and spoken-language understanding.

Previously, I graduated Summa Cum Laude from UC San Diego with a B.S. in Electrical Engineering and a Minor in Mathematics. I was born and raised in Nagoya, Japan until I graduated from High School. My Japanese name is 徐安峰 / Yasumine Jo.

Updates:

[Summer 2026] Started an internship at Meta Superintelligence Labs, voice applied team in Menlo Park.

[Summer 2025] Finished an internship at Meta Superintelligence Labs, voice modeling team in Seattle.

[Summer 2024] Finished an internship at Meta Reality Labs, wearable speech AI team in Seattle.

Last Updated: June 2026

Selected Publications

TASLP

End-to-End Joint ASR and Speaker Role Diarization with Child-Adult Interactions

Anfeng Xu, Tiantian Feng, Somer Bishop, and 2 more authors

IEEE Transactions on Audio, Speech and Language Processing (TASLP), 2026

Bib Link

@article{xu2026end,
  title = {End-to-End Joint ASR and Speaker Role Diarization with Child-Adult Interactions},
  author = {Xu, Anfeng and Feng, Tiantian and Bishop, Somer and Lord, Catherine and Narayanan, Shrikanth},
  journal = {IEEE Transactions on Audio, Speech and Language Processing (TASLP)},
  year = {2026}
}

Interspeech

Large Language Models based ASR Error Correction for Child Conversations

Anfeng Xu, Tiantian Feng, So Hyun Kim, and 3 more authors

Interspeech, 2025

Bib Link

@article{xu2025large,
  title = {Large Language Models based ASR Error Correction for Child Conversations},
  author = {Xu, Anfeng and Feng, Tiantian and Kim, So Hyun and Bishop, Somer and Lord, Catherine and Narayanan, Shrikanth},
  journal = {Interspeech},
  year = {2025}
}

ICASSP

Effective Integration of KAN for Keyword Spotting

Anfeng Xu, Biqiao Zhang, Shuyu Kong, and 4 more authors

ICASSP, 2025

Bib Link

@article{xu2024effective,
  title = {Effective Integration of KAN for Keyword Spotting},
  author = {Xu, Anfeng and Zhang, Biqiao and Kong, Shuyu and Huang, Yiteng and Yang, Zhaojun and Srivastava, Sangeeta and Sun, Ming},
  journal = {ICASSP},
  year = {2025}
}

ICASSP

Data efficient child-adult speaker diarization with simulated conversations

Anfeng Xu, Tiantian Feng, Helen Tager-Flusberg, and 2 more authors

ICASSP, 2025

Bib Link

@article{xu2024data,
  title = {Data efficient child-adult speaker diarization with simulated conversations},
  author = {Xu, Anfeng and Feng, Tiantian and Tager-Flusberg, Helen and Lord, Catherine and Narayanan, Shrikanth},
  journal = {ICASSP},
  year = {2025}
}

Interspeech

Exploring Speech Foundation Models for Speaker Diarization in Child-Adult Dyadic Interactions

Anfeng Xu, Kevin Huang, Tiantian Feng, and 3 more authors

Interspeech, 2024

Bib Link

@article{xu2024exploring,
  title = {Exploring Speech Foundation Models for Speaker Diarization in Child-Adult Dyadic Interactions},
  author = {Xu, Anfeng and Huang, Kevin and Feng, Tiantian and Shen, Lue and Tager-Flusberg, Helen and Narayanan, Shrikanth},
  journal = {Interspeech},
  year = {2024}
}

ICASSP

Audio-visual child-adult speaker classification in dyadic interactions

Anfeng Xu, Kevin Huang, Tiantian Feng, and 2 more authors

ICASSP, 2024

Bib Link

@article{xu2023audio,
  title = {Audio-visual child-adult speaker classification in dyadic interactions},
  author = {Xu, Anfeng and Huang, Kevin and Feng, Tiantian and Tager-Flusberg, Helen and Narayanan, Shrikanth},
  journal = {ICASSP},
  year = {2024}
}

Interspeech

Understanding Spoken Language Development of Children with ASD Using Pre-trained Speech Embeddings

Anfeng Xu, Rajat Hebbar, Rimita Lahiri, and 5 more authors

Interspeech, 2023

Bib Link

@article{xu2023understanding,
  title = {Understanding Spoken Language Development of Children with ASD Using Pre-trained Speech Embeddings},
  author = {Xu, Anfeng and Hebbar, Rajat and Lahiri, Rimita and Feng, Tiantian and Butler, Lindsay and Shen, Lue and Tager-Flusberg, Helen and Narayanan, Shrikanth},
  journal = {Interspeech},
  year = {2023}
}

CV:

Selected Publications