News

Sep 09, 2026 We are proud to announce that Phone Segmentation and Recognition through Phonological Activation Mapping, or SPAM (S3M-based Phonological Activation Mapping) for short, has been accepted to SLT 2026!
Phonetic structure is already present in the representations of self-supervised speech models (S3Ms), and we show that they can be steered to solve phone segmentation and recognition at the same time. The pipeline is simple and intuitive: S3M representations, phonological feature activations, and two gradient-descent-free prediction heads. Our method requires less than a minute of phonetic transcriptions and generalizes well to phones unseen during training!
Apr 07, 2026 The papers
  • POWSM: A Phonetic Open Whisper-Style Speech Foundation Model (main, top 5%)
  • PRiSM: Benchmarking Phone Realization in Speech Models (main)
  • Communicating in Emergent Language with an Induced Morphological Phrasebook (main)
  • [b] = [d] - [t] + [p]: Self-supervised Speech Models Discover Phonological Vector Arithmetic (findings)
  • Linear Script Representations in Speech Foundation Models Enable Zero-Shot Transliteration (findings)
  • PBEBench: A Multi-Step Programming by Examples Reasoning Benchmark inspired by Historical Linguistics (findings)
were accepted to ACL 2026.
Oct 12, 2025 David will give the Colloquium talk at LTI, “The Reconstruction Will Not Be Supervised.”
Sep 20, 2025 Changeling Lab member Brendon Boldt will present two papers in the main session of EMNLP 2025 (in Suzhou): “Morpheme Induction for Emergent Language” and “Searching for the Most Human-like Emergent Language.”
Sep 01, 2025 David Mortensen, Shinji Watanabe, and Jonathan Amith have received an NSF grant to leverage systematic patterns among related languages and dialects to improve ASR for low-resource varieties.
Aug 17, 2025 Chin-jou, Eunjung, Kwanghee, and David’s paper “Towards Inclusive ASR: Investigating Voice Conversion for Dysarthric Speech Recognition in Low-Resource Languages” was accepted to Interspeech 2025.
Aug 08, 2025 Atharva Naik, Kexun Zhang, Nate, Aravind Mysore, Clayton Marr, Hong Sng, Rebecca Byrnes, Anna, Kalvin, and David released a pre-print “Can Large Language Models Code Like a Linguist?: A Case Study in Low Resource Sound Law Induction.”
Jul 28, 2025 Eunjung and David’s journal paper “Applications of Artificial Intelligence for Cross-language Intelligibility Assessment of Dysarthric Speech” was accepted to Perspectives of the ASHA SIG 19.
Jul 01, 2025
  • “Programming by Example meets Historical Linguistics: A Large Language Model Based Approach to Sound Law Induction”
  • “DialUp! Modeling the Language Continuum by Adapting Models to Dialects and Dialects to Models” (Nate and David’s collaboration with Niyati Bafna, Emily Chang, Kenton Murray, David Yarowsky, and Hale Sirin)
  • “ZIPA: A family of efficient models for multilingual phone recognition” (collaboration with Jian Zhu, Farhan Samir, Eleanor Chodroff) were accepted to ACL 2025 (main).
Apr 29, 2025 Kwanghee, Eunjung, Kalvin, and David’s paper “Leveraging Allophony in Self-Supervised Speech Models for Atypical Pronunciation Assessment” was accepted to NAACL 2025 (main).
Apr 15, 2025 David gave a keynote on computational historical linguistics at Midwest Speech and Language Days 2025.
Feb 18, 2025 “Derivational morphology reveals analogical generalization in large language models” (Leonie and David’s collaboration with Valentin Hofmann, Hinrich Schütze, and Janet B. Pierrehumbert) was accepted to Proceedings of the National Academy of Sciences.
Jan 17, 2025 David Mortensen will give a talk as part of the University of Pittsburgh colloquium series.
Nov 01, 2024 Our paper “Zero-Shot Cross-Lingual NER Using Phonemic Representations for Low-Resource Languages” was accepted to EMNLP 2024 (main) and “Mitigating the Linguistic Gap with Phonemic Representations for Robust Cross-lingual Transfer” was accepted to MRL 2024.
Sep 10, 2024 Congratulations to Kalvin Chang and David Mortensen for winning an Honorable Mention, Best Paper Award at the Interspeech 2024 Responsible Speech Foundation Models Special Session with their paper “Self-supervised Speech Representations Still Struggle with African American Vernacular English” (joint work with Yi-Hui Chou, Hsuan-Ming Chen, Jiatong Shi, Nicole Holliday, and Odette Scharenborg).
Sep 06, 2024 David Mortensen gave a CLSP Seminar at Johns Hopkins University.
Aug 14, 2024 Congratulations to Liang (Leon) Lu for winning the ACL2024 Best Paper Award (Non-Publicized) with his paper “Semisupervised Neural Proto-language Reconstruction” (joint work with Peirong Xie and David Mortensen).
Aug 14, 2024
  • Kwanghee, Nate, and David’s paper “Wav2Gloss: Generating Interlinear Glossed Text from Speech” (collaboration with Taiqi He, Lindia Tjuatja, Jiatong Shi, Shinji Watanabe, Graham Neubig, Lori Levin)
  • Leon and David’s paper “Semisupervised Neural Proto-language Reconstruction” (collaboration with Peirong Xie) were accepted to ACL 2024.
Aug 07, 2024 :sparkles: ChangeLing Lab is officially born! :sparkles: (though it has long existed in fact).