Shikhar Bharadwaj
Logo PhD Student @ Carnegie Mellon University

Hi! I'm a PhD student at the Language Technologies Institute, Carnegie Mellon University, fortunate to be advised by Prof. David Mortensen and Prof. Shinji Watanabe. My research focuses on self-supervised learning and spoken language processing. I'm especially interested in using speech to learn world knowledge. This summer, I am a Research Scientist Intern at Meta Reality Labs.

Previously, I was a pre-doctoral researcher at Google DeepMind, working on low-resource language capabilities for Gemini with Dr. Partha Talukdar, Dr. Sriram Ganapathy and Dr. Shikhar Vashishth.

Besides my research, I find joy in playing snooker and freestyle football (soccer). I'm always open to collaborations, feel free to reach out!

Curriculum Vitae
Shikhar Bharadwaj
Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA
Email Google Scholar GitHub
Education
  • Carnegie Mellon University
    Carnegie Mellon University
    Language Technologies Institute
    Ph.D. Student
    Aug. 2024 - Present
  • Indian Institute of Science
    Indian Institute of Science
    Intelligent Systems
    M.Tech (Research)
    Aug. 2019 - Jun. 2022
  • BITS Pilani, Hyderabad
    BITS Pilani, Hyderabad
    Computer Science
    B.E. (Honors)
    Aug. 2014 - Jun. 2018
Experience
  • Meta Reality Labs
    Meta Reality Labs
    Research Scientist Intern
    May 2026 - Aug. 2026
  • Mitsubishi Electric Research Laboratories
    Mitsubishi Electric Research Laboratories
    PhD Research Intern
    May 2025 - Aug. 2025
  • Google DeepMind
    Google DeepMind
    Pre-Doctoral Researcher
    May 2022 - Jul. 2024
  • Microsoft Research
    Microsoft Research
    Research Intern
    Dec. 2017 - May 2018
News
2026
🎉 An Empirical Recipe for Universal Phone Recognition is accepted at Interspeech 2026 as an oral presentation!
Jun 18
🚀 Started as a Research Scientist Intern at Meta Reality Labs for Summer 2026.
May 18
💎 PRiSM and 🐁 POWSM (oral, top 5%) are accepted at ACL 2026!
Apr 15
2025
🎧 OpenBEATs, our fully open-source general-purpose audio encoder, is accepted at WASPAA 2025 as an oral presentation!
Jul 18
🎉 OpusLM is accepted at Interspeech 2025.
May 19

Selected Work (* equal contribution)

Phone Segmentation and Recognition through Phonological Activation Mapping

Shikhar Bharadwaj*, Kwanghee Choi*, Stephen McIntosh*, Chin-Jou Li, Eunjung Yeo, Daisuke Saito, Nobuaki Minematsu, Shinji Watanabe, Jian Zhu, David Harwath, David Mortensen

arXiv preprint 2026

[Paper]

Phone Segmentation and Recognition through Phonological Activation Mapping
Phone Segmentation and Recognition through Phonological Activation Mapping

Shikhar Bharadwaj*, Kwanghee Choi*, Stephen McIntosh*, Chin-Jou Li, Eunjung Yeo, Daisuke Saito, Nobuaki Minematsu, Shinji Watanabe, Jian Zhu, David Harwath, David Mortensen

arXiv preprint 2026

[Paper]

An Empirical Recipe for Universal Phone Recognition

Shikhar Bharadwaj, Chin-Jou Li, Kwanghee Choi, Eunjung Yeo, William Chen, Shinji Watanabe, David Mortensen

Interspeech (Oral) 2026

[Paper] [Code] [Checkpoint] [Demo]

An Empirical Recipe for Universal Phone Recognition
An Empirical Recipe for Universal Phone Recognition

Shikhar Bharadwaj, Chin-Jou Li, Kwanghee Choi, Eunjung Yeo, William Chen, Shinji Watanabe, David Mortensen

Interspeech (Oral) 2026

[Paper] [Code] [Checkpoint] [Demo]

PRiSM: Benchmarking Phone Realization in Speech Models

Shikhar Bharadwaj*, Chin-Jou Li*, Yoonjae Kim*, Kwanghee Choi, Eunjung Yeo, Ryan Shim, Hanyu Zhou, Brendon Boldt, Karen Jacome, Kalvin Chang, Darsh Agrawal, Keer Xu, Huck Yang, Jian Zhu, Shinji Watanabe, David Mortensen

Annual Meeting of the Association for Computational Linguistics (ACL) 2026

[Paper] [Code] [HuggingFace]

PRiSM: Benchmarking Phone Realization in Speech Models
PRiSM: Benchmarking Phone Realization in Speech Models

Shikhar Bharadwaj*, Chin-Jou Li*, Yoonjae Kim*, Kwanghee Choi, Eunjung Yeo, Ryan Shim, Hanyu Zhou, Brendon Boldt, Karen Jacome, Kalvin Chang, Darsh Agrawal, Keer Xu, Huck Yang, Jian Zhu, Shinji Watanabe, David Mortensen

Annual Meeting of the Association for Computational Linguistics (ACL) 2026

[Paper] [Code] [HuggingFace]

The CMU-AIST Submission for the ICME 2025 Audio Encoder Challenge

Shikhar Bharadwaj*, Samuele Cornell*, Kwanghee Choi, Hye-jin Shim, Soham Deshmukh, Satoru Fukayama, Shinji Watanabe

Technical Report (3rd place overall, top self-supervised model) 2026

[Paper] [ckpt1] [ckpt2]

The CMU-AIST Submission for the ICME 2025 Audio Encoder Challenge
The CMU-AIST Submission for the ICME 2025 Audio Encoder Challenge

Shikhar Bharadwaj*, Samuele Cornell*, Kwanghee Choi, Hye-jin Shim, Soham Deshmukh, Satoru Fukayama, Shinji Watanabe

Technical Report (3rd place overall, top self-supervised model) 2026

[Paper] [ckpt1] [ckpt2]

POWSM: A Phonetic Open Whisper-Style Speech Foundation Model

Chin-Jou Li*, Kalvin Chang*, Shikhar Bharadwaj, Eunjung Yeo, Kwanghee Choi, Jian Zhu, David Mortensen, Shinji Watanabe

Annual Meeting of the Association for Computational Linguistics (ACL, Oral, Top 5%) 2026

[Paper] [Code] [HuggingFace]

POWSM: A Phonetic Open Whisper-Style Speech Foundation Model
POWSM: A Phonetic Open Whisper-Style Speech Foundation Model

Chin-Jou Li*, Kalvin Chang*, Shikhar Bharadwaj, Eunjung Yeo, Kwanghee Choi, Jian Zhu, David Mortensen, Shinji Watanabe

Annual Meeting of the Association for Computational Linguistics (ACL, Oral, Top 5%) 2026

[Paper] [Code] [HuggingFace]

OpenBEATs: A Fully Open-Source General-Purpose Audio Encoder

Shikhar Bharadwaj, Samuele Cornell, Kwanghee Choi, Satoru Fukayama, Hye-jin Shim, Soham Deshmukh, Shinji Watanabe

IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA, Oral) 2025

[Paper] [Code] [HuggingFace]

OpenBEATs: A Fully Open-Source General-Purpose Audio Encoder
OpenBEATs: A Fully Open-Source General-Purpose Audio Encoder

Shikhar Bharadwaj, Samuele Cornell, Kwanghee Choi, Satoru Fukayama, Hye-jin Shim, Soham Deshmukh, Shinji Watanabe

IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA, Oral) 2025

[Paper] [Code] [HuggingFace]

Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

Gemini Team, Google (incl. Shikhar Bharadwaj)

Technical Report, Google DeepMind 2025

[Paper]

Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

Gemini Team, Google (incl. Shikhar Bharadwaj)

Technical Report, Google DeepMind 2025

[Paper]

OpusLM: A Family of Open Unified Speech Language Models

Jinchuan Tian, William Chen, Yifan Peng, Jiatong Shi, Siddhant Arora, Shikhar Bharadwaj, Takashi Maekaku, Yusuke Shinohara, Keita Goto, Xiang Yue, Huck Yang, Shinji Watanabe

Interspeech 2025

[Paper]

OpusLM: A Family of Open Unified Speech Language Models
OpusLM: A Family of Open Unified Speech Language Models

Jinchuan Tian, William Chen, Yifan Peng, Jiatong Shi, Siddhant Arora, Shikhar Bharadwaj, Takashi Maekaku, Yusuke Shinohara, Keita Goto, Xiang Yue, Huck Yang, Shinji Watanabe

Interspeech 2025

[Paper]

STAB: Speech Tokenizer Assessment Benchmark

Shikhar Vashishth*, Harman Singh*, Shikhar Bharadwaj*, Sriram Ganapathy, Chulayuth Asawaroengchai, Kartik Audhkhasi, Andrew Rosenberg, Ankur Bapna, Bhuvana Ramabhadran

arXiv preprint 2024

[Paper]

STAB: Speech Tokenizer Assessment Benchmark
STAB: Speech Tokenizer Assessment Benchmark

Shikhar Vashishth*, Harman Singh*, Shikhar Bharadwaj*, Sriram Ganapathy, Chulayuth Asawaroengchai, Kartik Audhkhasi, Andrew Rosenberg, Ankur Bapna, Bhuvana Ramabhadran

arXiv preprint 2024

[Paper]

IndicGenBench: A Multilingual Benchmark to Evaluate Generation Capabilities of LLMs on Indic Languages

Harman Singh, Nitish Gupta, Shikhar Bharadwaj, Dinesh Tewari, Partha Talukdar

Annual Meeting of the Association for Computational Linguistics (ACL) 2024

[Paper]

IndicGenBench: A Multilingual Benchmark to Evaluate Generation Capabilities of LLMs on Indic Languages
IndicGenBench: A Multilingual Benchmark to Evaluate Generation Capabilities of LLMs on Indic Languages

Harman Singh, Nitish Gupta, Shikhar Bharadwaj, Dinesh Tewari, Partha Talukdar

Annual Meeting of the Association for Computational Linguistics (ACL) 2024

[Paper]

Multimodal Modeling for Spoken Language Identification

Shikhar Bharadwaj*, Min Ma*, Shikhar Vashishth*, Ankur Bapna, Sriram Ganapathy, Vera Axelrod, Siddharth Dalmia, Wei Han, Yu Zhang, Daan van Esch, Sandy Ritchie, Partha Talukdar, Jason Riesa

IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2024

[Paper]

Multimodal Modeling for Spoken Language Identification
Multimodal Modeling for Spoken Language Identification

Shikhar Bharadwaj*, Min Ma*, Shikhar Vashishth*, Ankur Bapna, Sriram Ganapathy, Vera Axelrod, Siddharth Dalmia, Wei Han, Yu Zhang, Daan van Esch, Sandy Ritchie, Partha Talukdar, Jason Riesa

IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2024

[Paper]