#INTERSPEECH2024 で発表する主著論文「Spatial VC（ステレオVC）」をarXivに投稿しました！ 従来のVoice Conversion (VC)はシングルチャネルで処理していますが、「人間は2つの耳で空間的情報も処理してるんだからVCもマルチチャネルで空間情報を扱うべきじゃない？」という提案をする論文です。

ASJ1日目の午後のポスターセッションで、以下の内容で発表します！ 以前から取り組んでいる、サンプリング周波数非依存性の新たな評価手法を提案した、という内容です！ポスターなので、みなさんとたくさん議論できればと思っています！

Our paper entitled "Neural Analog Filter for Sampling-Frequency-Independent Convolutional Layer" has been accepted for APSIPA Trans. Signal & Info. Process. Congrats, Imamura-kun k_imamura !

Our paper titled "Neural Analog Filter for Sampling-Frequency-Independent Convolutional Layer" has been ACCEPTED for publication in APSIPA Trans. Info. Processing. Congratulations, Imamura-kun! 我々の「周波数非依存畳み込み層ニューラルアナログフィルタ」論文がアクセプトされました。

Our paper titled "Neural Analog Filter for Sampling-Frequency-Independent Convolutional Layer" has been accepted for publication in APSIPA Trans. Info. Processing. ジャーナル論文がアクセプトされました（初めてで嬉しい）！共著者の皆様ありがとうございました！

Our paper titled "Neural Analog Filter for Sampling-Frequency-Independent Convolutional Layer" is now available in APSIPA Trans. Info. Processing as follows. This is an open access paper. nowpublishers.com/article/Detail…

三月のSPEASIPで以下の発表をします！よろしくお願いします！

Early Accessで公開されてました👀 ieeexplore.ieee.org/document/11003…

Our paper entitled "Local Equivariance Error-Based Metrics For Evaluating Sampling-Frequency-Independent Property of Neural Network" has been accepted for #EUSIPCO2025. Congrats, Imamura-kun k_imamura !

EUSIPCO2025に主著がアクセプトされました！SFI音源分離の評価に関する内容です！ Local Equivariance Error-Based Metrics for Evaluating Sampling-Frequency-Independent Property of Neural Network

🚀 We just released MSR-UTMOS — a powerful model for speech quality prediction that supports 16kHz, 24kHz, and 48kHz audio! 🔍 Powered by a sampling frequency-independent convolutional layer on top of SSL models. 🎧 Upload your own samples and try it now:　huggingface.co/spaces/sarulab…

paper is available on arxiv arxiv.org/abs/2507.14647

Journal presentationで発表しました！ 論文の方も読んでいただけるととても嬉しいです🙇‍♂️ nowpublishers.com/article/Detail…

APSIPA翌日ですが、Oasisのライブ行ってきました！最高すぎました！

Our paper titled "Stride Conversion Algorithms for Convolutional Layers and Its Application to Sampling-Frequency-Independent Deep Neural Networks" has been ACCEPTED for publication in Signal Processing. Congratulations, Imamura-san!

Our paper titled "Stride Conversion Algorithms for Convolutional Layers and Its Application to Sampling-Frequency-Independent Deep Neural Networks Signal Processing" has been ACCEPTED for publication in Signal Processing (Elsevier). Congratulations, Imamura-kun!

My first-authored paper has been accepted for publication in Signal Processing (Elsevier)! "Stride Conversion Algorithms for Convolutional Layers and Its Application to Sampling-Frequency-Independent Deep Neural Networks''

This paper is now available! doi.org/10.1016/j.sigp…

Our paper titled “Dissecting Performance Degradation in Audio Source Separation under Sampling Frequency Mismatch” has been accepted for ICASSP 2026! Also, our journal paper (doi.org/10.1016/j.sigp…) has been accepted for journal presentation! See you in Balcelona!

Kanami Imamura, Tomohiko Nakamura, Kohei Yatabe, Hiroshi Saruwatari, "Dissecting Performance Degradation in Audio Source Separation under Sampling Frequency Mismatch," arxiv.org/abs/2601.14684