研究者詳細 - HALPERN Bence mark

写真a

ハルペン　ベンセ　マルク

HALPERN Bence mark

所属

情報基盤センターデータサイエンス研究部門特任助教

学位 2

Doctor (Linguistics) （ 2022年9月）
MEng (Biomedical Engineering) （ 2018年9月 Imperial College London ）

学位の先頭へ▲

研究キーワード 1

音声情報処理

研究キーワードの先頭へ▲

経歴 3

名古屋大学情報基盤センターデータサイエンス研究部門特任助教

2024年7月 - 現在

　詳細を見る

国名：日本国
名古屋大学大学院情報学研究科附属価値創造教育研究センター客員特任助教

2023年9月 - 2024年6月
Netherlands Cancer Institute Department of Head and Neck Oncology Researcher

2022年9月 - 2024年6月

　詳細を見る

国名：オランダ王国

経歴の先頭へ▲

所属学協会 2

ISCA

2022年9月 - 現在
International Speech Communication Association (ISCA) Member

所属学協会の先頭へ▲

受賞 1

SpandLDeteriorate@ACM MM Asia Best Paper Award

2024年12月

Bence Mark HALPERN, Tomoki TODA

受賞の先頭へ▲

論文 28

Severity-Controllable Pathological Text-to-Speech Synthesis for Clinical Applications 査読有り Open Access

Halpern, BM; Huang, WC; Violeta, LP; Toda, T

IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING 34 巻頁： 573 - 582 2026年

　詳細を見る

担当区分：筆頭著者,　責任著者掲載種別：研究論文（学術雑誌）出版者・発行元：IEEE Transactions on Neural Systems and Rehabilitation Engineering

The article presents a new pathological text-to-speech (TTS) synthesis system that has the ability to control speech severity using latent interpolations. Recognizing the difficulty of this task, our work uses a data augmentation technique to generate a single-speaker multi-severity training dataset required for training such a model. Furthermore, we show how x-vectors already contain information about the severity and leverage it as a conditioning variable for the synthesis. Finally, we propose modifications to the GradTTS architecture to enhance the duration modeling of pathological speech. We carry out objective and subjective evaluations to demonstrate that the proposed GradTTS system works well, and produces more natural, controllable, and stable pathological speech samples than the baseline TransformerTTS system.

添付ファイル： TNSRE3651761.pdf

DOI： 10.1109/TNSRE.2026.3651761

Open Access

Web of Science

Scopus

PubMed
Relationship between objective and subjective perceptual measures of speech in individuals with head and neck cancer

Bence Mark Halpern, Thomas Tienkamp, Teja Rebernik, Rob J.J.H. van Son, Martijn Wieling, Defne Abur, Tomoki Toda

Interspeech 2025年9月

　詳細を見る

担当区分：筆頭著者掲載種別：研究論文（国際会議プロシーディングス）

添付ファイル： B_11.pdf
XPPG-PCA: Reference-Free Automatic Speech Severity Evaluation With Principal Components 査読有り Open Access

Halpern, BM; Tienkamp, TB; Rebernik, T; van Son, RJJH; de Visscher, SAHJ; Witjes, MJH; Abur, D; Toda, T

IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING 19 巻 ( 5 ) 頁： 783 - 795 2025年7月

　詳細を見る

担当区分：筆頭著者掲載種別：研究論文（学術雑誌）出版者・発行元：IEEE Journal on Selected Topics in Signal Processing

Reliably evaluating the severity of a speech pathology is crucial in healthcare. However, the current reliance on expert evaluations by speech-language pathologists presents several challenges: while their assessments are highly skilled, they are also subjective, time-consuming, and costly, which can limit the reproducibility of clinical studies and place a strain on healthcare resources. While automated methods exist, they have significant drawbacks. Reference-based approaches require transcriptions or healthy speech samples, restricting them to read speech and limiting their applicability. Existing reference-free methods are also flawed; supervised models often learn spurious shortcuts from data, while handcrafted features are often unreliable and restricted to specific speech tasks. This paper introduces XPPG-PCA (x-vector phonetic posteriorgram principal component analysis), a novel, unsupervised, reference-free method for speech severity evaluation. Using three Dutch oral cancer datasets, we demonstrate that XPPG-PCA performs comparably to, or exceeds established reference-based methods. Our experiments confirm its robustness against data shortcuts and noise, showing its potential for real-world clinical use. Taken together, our results show that XPPG-PCA provides a robust, generalizable solution for the objective assessment of speech pathology, with the potential to significantly improve the efficiency and reliability of clinical evaluations across a range of disorders. An open-source implementation is available.

添付ファイル： XPPGPCA_Journal_Halpern_Proof_Final.pdf

DOI： 10.1109/JSTSP.2025.3617859

Web of Science

Scopus
Associations Between Acoustic, Kinematic, Self-Reported, and Perceptual Measures of Speech in Individuals Surgically Treated for Oral Cancer

Tienkamp, TB; Rebernik, T; Halpern, BM; van Son, RJJH; Wieling, M; Witjes, MJH; de Visscher, SAHJ; Abur, D

JOURNAL OF SPEECH LANGUAGE AND HEARING RESEARCH 68 巻 ( 7 ) 頁： 3069 - 3089 2025年7月

　詳細を見る

記述言語：英語掲載種別：研究論文（学術雑誌）

添付ファイル： A_6.pdf

DOI： 10.1044/2025_JSLHR-24-00464

Web of Science

PubMed
Reference-free automatic speech severity evaluation using acoustic unit language modelling 査読有り Open Access

Bence Mark Halpern, Tomoki Toda

Proceedings of the 6th ACM International Conference on Multimedia in Asia Workshops 頁： 1 - 5 2024年12月

　詳細を見る

DOI： https://doi.org/10.1145/3700410.3702114
Quantifying the effect of speech pathology on automatic and human speaker verification 査読有り国際共著 Open Access

Bence Mark Halpern, Thomas Tienkamp, Wen-Chin Huang, Lester Phillip Violeta, Teja Rebernik, Sebastiaan de Visscher, Max Witjes, Martijn Wieling, Defne Abur, Tomoki Toda

Interspeech 頁： 3015 - 3019 2024年9月

　詳細を見る

担当区分：筆頭著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

DOI： 10.21437/Interspeech.2024-1400

Open Access
Towards inclusive automatic speech recognition 査読有り国際共著 Open Access

Siyuan Feng, Bence Mark Halpern, Olya Kudina, Odette Scharenborg

Computer Speech & Language 84 巻 2024年3月

　詳細を見る

記述言語：英語掲載種別：研究論文（学術雑誌）

DOI： https://doi.org/10.1016/j.csl.2023.101567

Open Access
Towards inclusive automatic speech recognition 査読有り国際共著

Siyuan Feng, Bence Mark Halpern, Olya Kudina, Odette Scharenborg

Computer Speech & Language 84 巻 2024年3月

　詳細を見る

記述言語：英語掲載種別：研究論文（学術雑誌）

添付ファイル： A_5.pdf

DOI： 10.1016/j.csl.2023.101567
Quantifying Articulatory Working Space in Individuals Surgically Treated for Oral Cancer With Electromagnetic Articulography 査読有り国際共著

Quantifying Articulatory Working Space in Individuals Surgically Treated for Oral Cancer With Electromagnetic Articulography

Journal of Speech, Language, and Hearing Research 67 巻 ( 2 ) 頁： 384 - 399 2024年2月

　詳細を見る

記述言語：英語掲載種別：研究論文（学術雑誌）

DOI： 10.1044/2023_jslhr-23-00111
Quantifying Articulatory Working Space in Individuals Surgically Treated for Oral Cancer With Electromagnetic Articulography 査読有り国際共著

Quantifying Articulatory Working, Space in Individuals Surgically Treated for Oral Cancer With Electromagnetic Articulography

Journal of Speech, Language, and Hearing Research 67 巻 ( 2 ) 頁： 384 - 399 2024年2月

　詳細を見る

記述言語：英語掲載種別：研究論文（学術雑誌）

添付ファイル： A_4.pdf
Associations between acoustic, kinematic, self-reported and perceptual based measures of speech in individuals surgically treated for oral cancer 招待有り査読有り国際共著

Thomas Tienkamp*, Teja Rebernik, Bence Halpern, Rob van Son, Martijn Wieling, Max Witjes, Sebastiaan de Visscher, Defne Abur

2024 Motor Speech Conference 2024年2月

　詳細を見る

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）
Associations between acoustic, kinematic, self-reported and perceptual based measures of speech in individuals surgically treated for oral cancer 招待有り査読有り国際共著

Thomas Tienkamp, Teja Rebernik, Bence Halpern, Rob van Son, Martijn Wieling, Max Witjes, Sebastiaan de Visscher, Defne Abur

2024 Motor Speech Conference 2024年2月

　詳細を見る

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）
Reference-free automatic speech severity evaluation using acoustic unit language modelling 査読有り Open Access

Halpern, BM; Toda, T

PROCEEDINGS OF THE 6TH ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA IN ASIA WORKSHOPS, MMASIA 2024 WORKSHOPS 頁： 1 - 5 2024年

　詳細を見る

担当区分：筆頭著者,　責任著者出版者・発行元：Proceedings of the 6th ACM International Conference on Multimedia in Asia Workshops Mmasia 2024 Workshops

Speech severity evaluation is becoming increasingly important as the economic burden of speech disorders grows. Current speech severity models often struggle with generalization, learning dataset-specific acoustic cues rather than meaningful correlates of speech severity. Furthermore, many models require reference speech or a transcript, limiting their applicability in ecologically valid scenarios, such as spontaneous speech evaluation. Previous research indicated that automatic speech naturalness evaluation scores correlate strongly with severity evaluation scores, leading us to explore a reference-free method, SpeechLMScore, which does not rely on pathological speech data. Additionally, we present the NKI-SpeechRT dataset, based on the NKI-CCRT dataset, to provide a more comprehensive foundation for speech severity evaluation. This study evaluates whether SpeechLMScore outperforms traditional acoustic feature-based approaches and assesses the performance gap between reference-free and reference-based models. Moreover, we examine the impact of noise on these models by utilizing subjective noise ratings in the NKI-SpeechRT dataset. The results demonstrate that SpeechLMScore is robust to noise and offers superior performance compared to traditional approaches.

添付ファイル： B_10.pdf

DOI： 10.1145/3700410.3702114

Web of Science

Scopus
Improving severity preservation of healthy-to-pathological voice conversion with global style tokens 招待有り査読有り国際共著

Bence Mark Halpern, Wen-Chin Huang, Lester Phillip Violeta, RJJH van Son, Tomoki Toda

2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU) 頁： 1 - 7 2023年12月

　詳細を見る

担当区分：筆頭著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）
Automatic evaluation of spontaneous oral cancer speech using ratings from naive listeners 招待有り査読有り Open Access

Bence Mark Halpern, Siyuan Feng, Rob van Son, Michiel van den Brekel, Odette Scharenborg

Speech Communication 2023年3月

　詳細を見る

担当区分：筆頭著者,　責任著者

DOI： 10.1016/j.specom.2023.03.008

Open Access
Objective speech outcomes after surgical treatment for oral cancer: An acoustic analysis of a spontaneous speech corpus containing 32.850 tokens 査読有り Open Access

Thomas B Tienkamp, Rob JJH van Son, Bence Mark Halpern

Journal of Communication Disorders 2022年11月

　詳細を見る

担当区分：最終著者記述言語：英語掲載種別：研究論文（学術雑誌）

DOI： 10.1016/j.jcomdis.2022.106292

Open Access
Automatic Speech Recognition and Error Analyses of Dutch Oral Cancer Speech 招待有り査読有り

Kirsten Wildenburg, Bence M Halpern, Teja Rebernik, Thomas Tienkamp, Rob JJH van Son, Vass Verkhodanova, Max JH Witjes, Martijn Wieling

Young Female* Researchers in Speech Workshop (YFRSW) 2022年9月

　詳細を見る

記述言語：英語掲載種別：研究論文（研究会，シンポジウム資料等）
Quantifying changes in articulatory working space following oral cancer treatment 査読有り国際共著

Thomas Tienkamp, Teja Rebernik, Bence Halpern, Defne Abur, Rob van Son, Sebastiaan de Visscher, Max Witjes, Martijn Wieling

8th International Conference on Speech Motor Control: SMC2022 2022年8月

　詳細を見る

記述言語：英語
Low-resource automatic speech recognition and error analyses of oral cancer speech 査読有り Open Access

Bence Mark Halpern, Siyuan Feng, Rob van Son, Michiel van den Brekel, Odette Scharenborg

Speech Communication 141 巻頁： 14 - 27 2022年6月

　詳細を見る

担当区分：筆頭著者,　責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

DOI： 10.1016/j.specom.2022.04.006

Open Access
Towards identity preserving normal to dysarthric voice conversion 査読有り国際共著

Wen-Chin Huang, Bence Mark Halpern, Lester Phillip Violeta, Odette Scharenborg, Tomoki Toda

ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2022年5月

　詳細を見る

担当区分：筆頭著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）
The Effectiveness of Time Stretching for Enhancing Dysarthric Speech for Improved Dysarthric Speech Recognition 査読有り

Luke Prananta, Bence Mark Halpern, Siyuan Feng, Odette Scharenborg

Interspeech 2022 2022年1月

　詳細を見る

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）
Mitigating bias against non-native accents 査読有り国際共著

Yuanyuan Zhang, Yixuan Zhang, Bence Mark Halpern, Tanvina Patel, Odette Scharenborg

Interspeech 2022 頁： 3168 - 3172 2022年

　詳細を見る

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）
Manipulation of oral cancer speech using neural articulatory synthesis

Bence Mark Halpern, Teja Rebernik, Thomas Tienkamp, Rob van Son, Michiel van den Brekel, Martijn Wieling, Max Witjes, Odette Scharenborg

2022年

　詳細を見る

記述言語：英語掲載種別：研究論文（その他学術会議資料等）
An objective evaluation framework for pathological speech synthesis 査読有り国際共著

Bence Mark Halpern, Julian Fritsch, Enno Hermann, Rob van Son, Odette Scharenborg, Mathew Magimai Doss

ITG Speech Communications 2021 2021年7月

　詳細を見る

担当区分：筆頭著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）
Speaker-informed speech enhancement and separation 査読有り

Bence Mark Halpern, Finnian Kelly, Anil Alexander

IAFPA 2021年7月

　詳細を見る

担当区分：筆頭著者記述言語：英語掲載種別：研究論文（研究会，シンポジウム資料等）
Pathological voice adaptation with autoencoder-based voice conversion 招待有り査読有り

Marc Illa, Bence Mark Halpern, Rob van Son, Laureano Moro-Velazquez, Odette Scharenborg

Speech Synthesis Workshop 2021 2021年6月

　詳細を見る

担当区分：筆頭著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）
Quantifying Bias in Automatic Speech Recognition

Siyuan Feng, Olya Kudina, Bence Mark Halpern, Odette Scharenborg

2021年3月

　詳細を見る

記述言語：英語掲載種別：研究論文（その他学術会議資料等）
Detecting and analysing spontaneous oral cancer speech in the wild 招待有り査読有り Open Access

Bence Mark Halpern, Rob van Son, Michiel van den Brekel, Odette Scharenborg

Interspeech 2020 頁： 4826 - 4830 2020年7月

　詳細を見る

担当区分：筆頭著者,　責任著者

DOI： 10.21437/Interspeech.2020-1598

Open Access

▼全件表示

論文の先頭へ▲

共同研究・競争的資金等の研究課題 1

"I don't sound like myself": Creating voice conversion-based speech technology for healthcare

2024年7月 - 2026年7月

Voice conversion

　詳細を見る

担当区分：研究代表者資金種別：競争的資金

配分額：150円（直接経費：2600000円）

共同研究・競争的資金等の研究課題の先頭へ▲

科研費 1

"I don't sound like myself": Creating voice conversion-based speech technology for healthcare

2024年7月 - 2026年7月

Voice conversion

　詳細を見る

担当区分：研究代表者資金種別：競争的資金

配分額：150円（直接経費：2600000円）

科研費の先頭へ▲