IEICE Technical Report

Online edition: ISSN 2432-6380

Speech

Workshop Date : 2026-06-05 - 2026-06-06 / Issue Date : 2026-05-29

SP2026-1
[Poster Presentation] A Study on Data Augmentation for Speech Emotion Recognition Using Personality, Conversational Context and VAD Scores
Sho Yamakawa, Takiko Sasaki (Musashino Univ.)
pp. 1 - 6

SP2026-2
(See Japanese page.)
pp. 7 - 12

SP2026-3
(See Japanese page.)
pp. 13 - 18

SP2026-4
A study on laugher-to-speech and laughter-to-laughter separation using Conv-TasNet
Akihiro Miyake, Yoshiko Arimoto (Chiba Tech)
pp. 19 - 24

SP2026-5
[Poster Presentation] Reinforcement Learning with Domain-Specific Perceptual Reward for Anime Style-Targeted Text-to-Speech Optimization
Joonyong Park, Jerry Li (Spellbrush)
pp. 25 - 30

SP2026-6
[Poster Presentation] Identity-Preserving Speaker Attribute Editing
Yuta Seki, Koichi Shinoda (Science Tokyo)
pp. 31 - 36

SP2026-7
[Poster Presentation] A Study on Multimodal Dialog Act Classification Using Corpus of Everyday Japanese Conversation
Shu Makioka, Takeshi Yamada (Univ. Tsukuba)
pp. 37 - 42

SP2026-8
[Poster Presentation] Speech Emotion Recognition Based on Reconstruction of Latent Features in Self-Supervised Speech Models
Sora Hashimoto, Takeshi Yamada (Univ. Tsukuba)
pp. 43 - 48

SP2026-9
[Poster Presentation] Computation of dynamic time response of moving receiver
Yuto Oyama, Kunitoshi Motoki (Hokkai-Gakuen Univ.)
pp. 49 - 52

SP2026-10
[Poster Presentation] Real-time Voice Activity Detection for Speech Recognition in Dialogues between Known and Unknown Speakers
Keiko Yamamoto, Kosuke Osumi, Martin Klinkigt, Masanori Morobishi, Takayuki Arakawa (KYOCERA)
pp. 53 - 60

SP2026-11
[Poster Presentation] Robust Speaker Representation Learning via Self-Supervised Pre-training Using Diffusion Models
Kaito Miwa, Koichi Shinoda (Science Tokyo)
pp. 61 - 65

SP2026-12
How J-POP and K-POP are different
Naofumi Aoki (Hokkaido Univ.)
pp. 66 - 70

SP2026-13
[Poster Presentation] Separating peripheral hearing loss from higher-level effects in speech intelligibility of older adults
Toshio Irino (Wakayama Univ./ Univ. Electro-Comm.), Ayako Yamamoto, Fuki Miyazaki (Wakayama Univ.)
pp. 71 - 78

SP2026-14
A study on scream synthesis using latent diffusion model
Reiji Hirao, Yoshiko Arimoto (Chiba Tech)
pp. 79 - 83

SP2026-15
[Poster Presentation] Fundamental Frequency Estimation of Sound Signal with Missing F0 in Noisy Environments
Yoshimasa Tsuru, Kenichiro Miwa (Salesian Polytechnic)
pp. 84 - 86

SP2026-16
A Preliminary Study of Prosodic Self-Regulation Training Using Individual Baseline Feedback Based on a Three-Dimensional Attitude Impression Space
Chizu Sugai, Hirofumi Kikuchi, Hideaki Kikuchi (Waseda Univ.)
pp. 87 - 90

SP2026-17
[Poster Presentation] Visual Speech Recognition via Large Language Model Distillation and Viseme-Aware CTC Learning
Haruki Komai, Koichi Shinoda (Science Tokyo)
pp. 91 - 98

SP2026-18
Effects of Non-Invasive Facial Muscle Interventions on the Psychological States and Acoustic Characteristics of Speakers
Youko Noge, Anna Oura, Hideaki Kikuchi (Waseda Univ.)
pp. 99 - 103

SP2026-19
[Poster Presentation] Study on F0 Estimation in Historical Sound Sources
Kasho Nishino, Kenichiro Miwa (Salesian Polytechnic)
pp. 104 - 106

SP2026-20
(See Japanese page.)
pp. 107 - 112

SP2026-21
[Poster Presentation] Single-Channel Speech Separation with Pseudo-Spatial Embeddings
Daichi Nitsu, Koichi Shinoda (Science Tokyo)
pp. 113 - 118

SP2026-22
Audio-Visual Joint Separation and Tracking of Moving Sources Based on a Hierarchical Generative Model Using Distributed Microphone and Camera Arrays
Ryunosuke Nihei (Kyoto Univ.), Kazuyoshi Yoshii (AIST), Aditya Arie Nugraha (RIKEN), Yosuke Ito, Kazuyoshi Yoshii (Kyoto Univ.)
pp. 119 - 124

SP2026-23
[Poster Presentation] Cross-layer recurrent processing of Residual Vector Quantization layers for Audio Codec-based speech separation
Phuong Dinh Nhu Minh, Roland Hartanto, Koichi Shinoda (Science Tokyo)
pp. 125 - 129

SP2026-24
Multi-Session EEG-to-Speech Decoding Combining Contrastive Learning with Variational Regularization
Tomoaki Mizuno, Toru Nakashika (UEC)
pp. 130 - 135

SP2026-25
Revisiting Mode Dependence in the Zipformer for Unified Streaming and Offline ASR
Wen Shen Teo, Yasuhiro Minami (UEC)
pp. 136 - 141

SP2026-26
A research plan on demonstrating the language-independence of the mechanism of speech-laugh occurrence
Takuya Tanimura, Yoshiko Arimoto (Chiba Tech)
pp. 142 - 147

SP2026-27
A Preliminary Investigation of Diverse Big Five Personality-Based Speech Synthesis Using Mixture Density Networks
Souma Narikawa, Tomoaki Mizuno, Toru Nakashika (UEC)
pp. 148 - 153

SP2026-28
Comparative Study of Alignment Methods for Speech Synthesis from EEG
Tomohisa Sakaguchi, Tomoaki Mizuno, Toru Nakashika (UEC)
pp. 154 - 159

SP2026-29
Learning latent representations for diverse control of screams in speech synthesis
Syuusei Arima, Hiroki Mori (Utsunomiya Univ.)
pp. 160 - 162

Note: Each article is a technical report without peer review, and its polished version will be published elsewhere.

The Institute of Electronics, Information and Communication Engineers (IEICE), Japan