no code implementations • 12 Apr 2024 • Masahiro Yasuda, Noboru Harada, Yasunori Ohishi, Shoichiro Saito, Akira Nakayama, Nobutaka Ono
This is because the information obtained from a single sensor is often missing or fragmented in such an environment; observations from multiple locations and modalities should be integrated to analyze events comprehensively.
no code implementations • 4 Mar 2024 • Masahiro Yasuda, Shoichiro Saito, Akira Nakayama, Noboru Harada
A system trained only with a dataset using microphone arrays in a fixed position would be unable to adapt to the fast relative motion of sound events associated with self-motion, resulting in the degradation of SELD performance.
1 code implementation • 1 Mar 2023 • Noboru Harada, Daisuke Niizumi, Yasunori Ohishi, Daiki Takeuchi, Masahiro Yasuda
This paper provides a baseline system for First-shot-compliant unsupervised anomaly detection (ASD) for machine condition monitoring.
1 code implementation • 18 Feb 2022 • Masahiro Yasuda, Yasunori Ohishi, Shoichiro Saito, Noboru Harada
We tackle a challenging task: multi-view and multi-modal event detection that detects events in a wide-range real environment by utilizing data from distributed cameras and microphones and their weak labels.
1 code implementation • 18 Feb 2022 • Masahiro Yasuda, Yasunori Ohishi, Shoichiro Saito
Our goal is to develop a sound event localization and detection (SELD) system that works robustly in unknown environments.
1 code implementation • 17 Feb 2022 • Kento Nagatomo, Masahiro Yasuda, Kohei Yatabe, Shoichiro Saito, Yasuhiro Oikawa
Sound event localization and detection (SELD) is a combined task of identifying the sound event and its direction.
no code implementations • 16 Feb 2022 • Tomoro Tanaka, Kohei Yatabe, Masahiro Yasuda, Yasuhiro Oikawa
Still, they cannot perform well if the training data have mismatches and/or constraints in the time domain are not imposed.
7 code implementations • 4 Jun 2021 • Noboru Harada, Daisuke Niizumi, Daiki Takeuchi, Yasunori Ohishi, Masahiro Yasuda, Shoichiro Saito
This paper proposes a new large-scale dataset called "ToyADMOS2" for anomaly detection in machine operating sounds (ADMOS).
no code implementations • 14 Dec 2020 • Yuma Koizumi, Yasunori Ohishi, Daisuke Niizumi, Daiki Takeuchi, Masahiro Yasuda
Then, the caption of the audio input is generated by using a pre-trained language model while referring to the guidance captions.
no code implementations • 1 Jul 2020 • Yuma Koizumi, Ryo Masumura, Kyosuke Nishida, Masahiro Yasuda, Shoichiro Saito
TRACKE estimates keywords, which comprise a word set corresponding to audio events/scenes in the input audio, and generates the caption while referring to the estimated keywords to reduce word-selection indeterminacy.
3 code implementations • 10 Jun 2020 • Yuma Koizumi, Yohei Kawaguchi, Keisuke Imoto, Toshiki Nakamura, Yuki Nikaido, Ryo Tanabe, Harsh Purohit, Kaori Suefusa, Takashi Endo, Masahiro Yasuda, Noboru Harada
The main challenge of this task is to detect unknown anomalous sounds under the condition that only normal sound samples have been provided as training data.
no code implementations • 10 Oct 2019 • Masahiro Yasuda, Yuma Koizumi, Luca Mazzon, Shoichiro Saito, Hisashi Uematsu
We propose a direction of arrival (DOA) estimation method that combines sound-intensity vector (IV)-based DOA estimation and DNN-based denoising and dereverberation.
no code implementations • 10 Oct 2019 • Luca Mazzon, Yuma Koizumi, Masahiro Yasuda, Noboru Harada
The same transformation is applied also to the labels, in order to maintain consistency between input data and target labels.