Výsledky vyhledávání - "Motoi Omachi"

Toward Streaming ASR with Non-Autoregressive Insertion-Based Model

Autor: Tianzi Wang, Shinji Watanabe, Yuya Fujita, Motoi Omachi

Publikováno v: Interspeech 2021.

Neural end-to-end (E2E) models have become a promising technique to realize practical automatic speech recognition (ASR) systems. When realizing such a system, one important issue is the segmentation of audio to deal with streaming input or long reco

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6154c2683e7dfa35c0b913d87fa38b49
https://doi.org/10.21437/interspeech.2021-1131

Zobrazit plný text záznamu

End-to-End ASR with Adaptive Span Self-Attention

Autor: Xuankai Chang, Motoi Omachi, Aswin Shanmugam Subramanian, Shinji Watanabe, Yuya Fujita, Pengcheng Guo

Publikováno v: INTERSPEECH

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::a7a1ab1236a0e0e060275cf3e0b3dc1b
https://doi.org/10.21437/interspeech.2020-2816

Zobrazit plný text záznamu

Insertion-Based Modeling for End-to-End Automatic Speech Recognition

Autor: Shinji Watanabe, Motoi Omachi, Yuya Fujita, Xuankai Chang

Publikováno v: INTERSPEECH

End-to-end (E2E) models have gained attention in the research field of automatic speech recognition (ASR). Many E2E models proposed so far assume left-to-right autoregressive generation of an output token sequence except for connectionist temporal cl

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8cf152c1d9616b3ec8695a578fe39351
http://arxiv.org/abs/2005.13211

Zobrazit plný text záznamu

Associative Memory Model-Based Linear Filtering and Its Application to Tandem Connectionist Blind Source Separation

Autor: Tetsunori Kobayashi, Motoi Omachi, Tetsuji Ogawa

Publikováno v: IEEE/ACM Transactions on Audio, Speech, and Language Processing. 25:637-650

We propose a blind source separation method that yields high-quality speech with low distortion. Time-frequency (TF) masking can effectively reduce interference, but it produces nonlinear distortion. By contrast, linear filtering using a separation m

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::06fd6c315295ca2ab2345fd4d6a6d351
https://doi.org/10.1109/taslp.2017.2653941

Zobrazit plný text záznamu

Attention-based ASR with Lightweight and Dynamic Convolutions

Autor: Aswin Shanmugam Subramanian, Shinji Watanabe, Motoi Omachi, Yuya Fujita

Publikováno v: ICASSP

End-to-end (E2E) automatic speech recognition (ASR) with sequence-to-sequence models has gained attention because of its simple model training compared with conventional hidden Markov model based ASR. Recently, several studies report the state-of-the

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f42f277e4c7a13f3e27d78b8e032e6f3
http://arxiv.org/abs/1912.11793

Zobrazit plný text záznamu

Speaker Selective Beamformer with Keyword Mask Estimation

Autor: Dung Tran, Motoi Omachi, Yusuke Kida, Yuya Fujita, Toru Taniguchi

Publikováno v: SLT

This paper addresses the problem of automatic speech recognition (ASR) of a target speaker in background speech. The novelty of our approach is that we focus on a wakeup keyword, which is usually used for activating ASR systems like smart speakers. T

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2dd514056bd6319fe044c81a6619dff3
https://doi.org/10.1109/slt.2018.8639651

Zobrazit plný text záznamu

Multi Scale Feedback Connection for Noise Robust Acoustic Modeling

Autor: Yuya Fujita, Dung Tran, Ken-ichi Iso, Motoi Omachi

Publikováno v: ICASSP

Simply feeding of a last hidden layer of the deep neural network (DNN) back to the input layer recently found to be effective for noise robust acoustic modeling. Such high level feature strengthens the robustness of DNN based acoustic model while pay

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::d392d7b98712dfdc588bc5d8cbffcfdd
https://doi.org/10.1109/icassp.2018.8461602

Zobrazit plný text záznamu

Separation matrix optimization using associative memory model for blind source separation

Autor: Masaru Fujieda, Tetsuji Ogawa, Motoi Omachi, Tetsunori Kobayashi, Kazuhiro Katagiri

Publikováno v: EUSIPCO

A source signal is estimated using an associative memory model (AMM) and used for separation matrix optimization in linear blind source separation (BSS) to yield high quality and less distorted speech. Linear-filtering-based BSS, such as independent

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7615a7aa122672c9f033cac8b3bb5677

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání