Our paper Fine-tuning TitaNet-Large Model for Speaker Anonymization Attacker Systems has been presented at the 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025), Hyderabad, India.

The new fiscal year 2025 has begun. This year, I will be teaching course I225E Statistical Signal Processing with Professor Unoki in Term 1-1 (April-June). Wishing you all a productive year ahead!

I participated in the 2025 Spring meeting of the Acoustical Society of Japan (ASJ), Saitama University, Japan. Three of our reserach papers were presented in poster sessions.

Our paper Beyond Accuracy: Multimodal Modeling of Structured Speaking Skill Indices in Young Adolescents has been accepted for publication in Journal of Computers & Education: Artificial Intelligence.

I receive the JSPS Kakenhi for Early-Career Scientists FY2025 (科研費 若手研究)).

Mr. Hiroto Horimoto and Mr. Shinya Koshikawa, master students from Okada laboratory, finished their final defense. お疲れ様でした!

Mr. Wenqing Wei and Mr. Fuminori Nagasawa, Ph.D candidates from Okada laboratory, finished their final defense. お疲れ様でした!

Our paper Influence of Personality Traits and Demographics on Rapport Recognition Using Adversarial Learning has been accepted for publication in Multimodal Technologies and Interaction.

Our paper Speech Intelligibility Prediction Using Binaural Processing for Hearing Loss has been accepted for publication in IEEE Access.

Our paper WorldCuisines: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global Cuisines has been accepted for presentation at the 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL 2025), Albuquerque, New Mexico. Deep appreciation to all co-authors for their dedication to the success of this big project.

Our paper Indonesian Speech Content De-Identification in Low Resource Transcripts has been presented by our student, Mr. Rifqi Naufal Abdjul, at The Second Workshop in South East Asian Language Processing (SEALP 2025), Co-loacted with (COLING 2025), Abu Dhabi (Online).

Our paper Fine-tuning TitaNet-Large Model for Speaker Anonymization Attacker Systems has been accepted for presentation at the 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025) - Grand Challenge (The First VoicePrivacy Attacker Challenge), Hyderabad, India.

Cheers to a year of new adventures, fresh starts, and endless possibilities. Happy New Year 2025!

We are glad that our proposed method in The First VoicePrivacy Attacker Challenge has been selected as one of the best-5 systems!

Three of our papers Detecting Spoof Voices in Asian Non-Native Speech: An Indonesian and Thai Case Study, Unsupervised Anomalous Sound Detection Using Timbral and Human Voice Disorder-Related Acoustic Features, and Anomalous Sound Detection Based on Time Domain Gammatone Filterbank and IDNN Model have been presented at the 16th annual conference organized by Asia-Pacific Signal and Information Processing Association (APSIPA 2024).

Our paper Indonesian Speech Content De-Identification in Low Resource Transcripts has been accepted for presentation at The Second Workshop in South East Asian Language Processing (SEALP 2025), Co-loacted with (COLING 2025), Abu Dhabi, UAE. This paper was the collaboration work between ITB (Indonesia) and JAIST (Japan). This work was supported by the JST Sakura Science Exchange Program FY2023.

We're excited to announce the publication of our latest research in the Journal of the Audio Engineering Society. This research explores the intersection of forensic science and speech technology aiming to ensure reliable understanding of poor-quality speech recordings used as evidence in criminal trials.

We successfully organized the 3rd ASEAN-IVO meeting and JAIST-ASEAN deepfake detection hub symposium. We welcomed participants from NICT, Japan and ASEAN institutions from Thailand, Indonesia, Brunei Darussalam, and Myanmar. The meetings were a resounding success, generating excitement and enthusiasm among all participants. The collaborative discussions and knowledge sharing fostered a vibrant atmosphere, laying the groundwork for future collaborations in spoof detection research.

Our paper Indonesian Speech Anti-Spoofing System: Data Creation and Convolutional Neural Network Models has been presented by our student, Ms. Sarah Azka Arief, at the 11th International Conference on Advanced Informatics: Concepts, Theory, and Applications (ICAICTA 2024), Singapore.

Our paper MAG-BERT-ARL for Fair Automated Video Interview Assessment has been accepted for publication in IEEE Access. This paper was the collaboration work between UI (Indonesia) and JAIST (Japan). This work was supported by the JST Sakura Science Exchange Program FY2023.

Our papers Unsupervised Anomalous Sound Detection Using Timbral and Human Voice Disorder-Related Acoustic Features and Anomalous Sound Detection Based on Time Domain Gammatone Filterbank and IDNN Model have been accepted for presentation at the 16th annual conference organized by Asia-Pacific Signal and Information Processing Association (APSIPA 2024). These papers were the collaboration work between ITB (Indonesia) and JAIST (Japan). This work was supported by the JST Sakura Science Exchange Program FY2023.

Our paper Detecting Spoof Voices in Asian Non-Native Speech: An Indonesian and Thai Case Study has been accepted for presentation at the 16th annual conference organized by Asia-Pacific Signal and Information Processing Association (APSIPA 2024).

Our papers UCSYSpoof: A Myanmar Language Dataset for Voice Spoofing Detection and Analysis of Pathological Features for Spoof Detection have been accepted for presentation at the 27th International Conference of the Oriental COCOSDA. These papers were the collaboration work between UCSY (Myanmar), NECTEC (Thailand), and JAIST (Japan). They were also a part of the ASEAN IVO project titled ‘Spoof Detection for Automatic Speaker Verification’(www.nict.go.jp/en/asean_ivo).

I participated in the 4th Symposium on Security and Privacy in Speech Communication and 3rd VoicePrivacy Challenge, Kos Island, Greece. This event is a satellite event of Interspeech. I served as a member of program committee and one of session chairs during the VPC presentations.

I participated in the Interspeech 2024, Kos Island, Greece. Our paper Are Recent Deep Learning-Based Speech Enhancement Methods Ready to Confront Real-World Noisy Environments? was presented as a poster presentation and received a lot of visitors from academia, industry, and others. I am very glad to be able to participate in this conference on-site for the first time!

Our paper Indonesian Speech Anti-Spoofing System: Data Creation and CNN Models has been accepted for presentation at the 11th International Conference on Advanced Informatics: Concepts, Theory, and Applications (ICAICTA 2024). This paper was the collaboration work between ITB (Indonesia) and JAIST (Japan).