Result for "speech recognition",271

Multilingual Speech Recognition Platform - Dual Channel Speech Recognition System｜NTT R&D Website

Multilingual Speech Recognition Platform - Dual Channel Speech Recognition System｜NTT R&D Website

https://www.rd.ntt/e/research/HI0002.html

Speech Recognition for Computers | NTT Communication Science Laboratories | NTT R&D Website

Speech Recognition for Computers | NTT Communication Science Laboratories | NTT R&D Website NTT R

https://www.rd.ntt/e/cs/team_project/media/signal/research_media05.html

Speech Recognition Technologies for Creating Knowledge Resources from Communications｜NTT R&D Website

Speech Recognition Technologies for Creating Knowledge Resources from Communications｜NTT R&D

https://www.rd.ntt/e/research/JN202010_7482.html

Tomohiro Nakatani | NTT R&D Website

/center/dept Audio and speech signal processing for capturing and recognition of human conversations The

https://www.rd.ntt/e/organization/researcher/superior/s_012.html

poster_en_21.pdf

our speech processing technologies developed for conversational speech recognition. Specifically, our

https://www.rd.ntt/cs/event/openhouse/2017/exhibition/21/poster_en_21.pdf

NTT Communication Science Laboratories Open House 2014 Defeat reverberation: enemy of speech recognition - Advanced speech enhancement and recognition -

recognition - Advanced speech enhancement and recognition - 日本語 English Home Greeting Program Presentations

https://www.rd.ntt/cs/event/openhouse/2014/exhibition/20/index_en.html

スライド 1

Learning speech recognition from small paired data - Semi-supervised end-to-end training with text-to

https://www.rd.ntt/cs/event/openhouse/2019/download/16_c_en.pdf

poster_en_20.pdf

speech recognition is used more and more often in our everyday life. However, the accuracy of speech

https://www.rd.ntt/cs/event/openhouse/2017/exhibition/20/poster_en_20.pdf

NTT Communication Science Laboratories Open House 2016

and signal processing that are making speech recognition leap forward - Takuya Yoshioka, Media

https://www.rd.ntt/cs/event/openhouse/2016/talk/research1/index_en.html

poster_en_19.pdf

Speech Recognition: Support 10 languages in Asia-Pacific region, using DNN structure robust for noisy

https://www.rd.ntt/cs/event/openhouse/2017/exhibition/19/poster_en_19.pdf

Speech information processing | NTT R&D Website

processing technology handles speech as signal data for analysis and processing, such as recognition and

https://www.rd.ntt/e/hil/category/voice/

中谷 智広 | NTT R&D Website

Recognition and Understanding Workshop (ASRU) Best Paper Award Honarable Mention 2015年　3rd CHiME Speech

https://www.rd.ntt/organization/researcher/superior/s_012.html

NTT Communication Science Laboratories Open House 2014 How accurate are speech recognition results? - Estimating speech recognition accuracy without references -

NTT Communication Science Laboratories Open House 2014 How accurate are speech recognition results

https://www.rd.ntt/cs/event/openhouse/2014/exhibition/21/index_en.html

poster_en.pdf

lectures by using speech recognition and machine translation technology. The system allows non-native

https://www.rd.ntt/cs/event/openhouse/2016/exhibition/23/poster_en.pdf

WPE speech dereverberation

with advanced speech enhancement and recognition technologies for the REVERB challenge," in Proceedings

https://www.rd.ntt/cs/team_project/media/signal/wpe/references.html

NTT Communication Science Laboratories Open House 2019

Learning speech recognition from small paired data - Semi-supervised end-to-end training with text-to

https://www.rd.ntt/cs/event/openhouse/2019/exhibition16/index_en.html

Microsoft Word - IWANNCE33_final.doc

Energy Operator in advancing the noise robustness of speech recognition front-end. The presented method

https://www.rd.ntt/cs/team_project/icl/signal/iwaenc03/cdrom/data/0033.pdf

NTT Communication Science Laboratories Open House 2017

Exhibition Program 21 Turn-taking matters in conversation recognition Robust speech processing using speakers

https://www.rd.ntt/cs/event/openhouse/2017/exhibition/21/index_en.html

Microsoft Word - paper0048.doc

presented. The method is effective for automatic speech recognition under conditions of multiple sound

https://www.rd.ntt/cs/team_project/icl/signal/iwaenc03/cdrom/data/0048.pdf

Marc Delcroix | NTT R&D Website

2015 Best Paper Award Honorable Mention at the 2015 IEEE Automatic Speech Recognition and Understanding

https://www.rd.ntt/e/organization/researcher/special/s_033.html

音声情報処理 | NTT R&D Website

, Marc Delcroix, Takahiro Shinozaki, "Streaming End-to-End Target-Speaker Automatic Speech Recognition

https://www.rd.ntt/hil/category/voice/

NTT Communication Science Laboratories Open House 2017

recognition Abstract Automatic speech recognition is used more and more often in our everyday life. However

https://www.rd.ntt/cs/event/openhouse/2017/exhibition/20/index_en.html

NTT Communication Science Laboratories Open House 2015

Extracting essential information from sounds - Advances in distant speech recognition by deep learning

https://www.rd.ntt/cs/event/openhouse/2015/exhibition/20/index_en.html

Signal Processing Research Group | NTT Communication Science Laboratories | NTT R&D Website

Digital Signal Processing technology, we are researching speech recognition technology, acoustic signal

https://www.rd.ntt/e/cs/team_project/media/signal/

pp.dvi

-independent system. The first assumption, domi- nant in the speech and speaker recognition literature todate

https://www.rd.ntt/cs/team_project/icl/signal/ica2003/cdrom/data/0177.pdf

D01-03-e.pdf

speech recognition technology for instant information grasp and decision-making support #Productivity

https://www.rd.ntt/forum/2024/doc/D01-03-e.pdf

poster_en18.pdf

and formally uttered. However, the recognition of casual speech is still a challenging task since

https://www.rd.ntt/cs/event/openhouse/2013/exhibition/media6/poster_en18.pdf

信号処理研究グループ｜NTTコミュニケーション科学基礎研究所｜NTT R&D Website

Speech Enhancement Errors on Speech Recognition Performance. IEEE/ACM Transactions on Audio, Speech, and

https://www.rd.ntt/cs/team_project/media/signal/

NTT Communication Science Laboratories Open House 2013 Transcribing every word whoever speaks - Speech recognizers robust to casual speech variations -

speech recognition is considered to be a mature technology that can accurately recognize sentences that

https://www.rd.ntt/cs/event/openhouse/2013/exhibition/media6/index_en.html

"Speech Enhancement based on Linear Prediction Error Signals and Spectral Subtraction"

@pino.datsi.fi.upm.es Abstract Speech processing and recognition are key technologies to produce smart

https://www.rd.ntt/cs/team_project/icl/signal/iwaenc03/cdrom/data/0077.pdf

iwaenc03-bp.dvi

front-end of the speech recognition system was confirmed. However, the model estimates direction of

https://www.rd.ntt/cs/team_project/icl/signal/iwaenc03/cdrom/data/0066.pdf

Media Information Laboratory | NTT Communication Science Laboratories | NTT R&D Website

, "Frontend Token Enhancement for Token-Based Speech Recognition" ・Y. Ren, N. Harada, D. Takeuchi, S. Chen, W

https://www.rd.ntt/e/cs/team_project/media/

0062.pdf

accuracy of an Automatic Speech Recognition (ASR) in comparison to other noise reduction methods. 1

https://www.rd.ntt/cs/team_project/icl/signal/iwaenc03/cdrom/data/0062.pdf

NTT Communication Science Laboratories Open House 2013

multi-speaker conversational speech recognition: past, present and future - Takaaki Hori, Media

https://www.rd.ntt/cs/event/openhouse/2013/talk/research1/index_en.html

Media Information Laboratory Past news | NTT Communication Science Laboratories | NTT R&D Website

Award at the 2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU). November 2025 The

https://www.rd.ntt/e/cs/team_project/media/past_news.html

スライド 1

, which offers a way to control the vocal expression of emotion in speech through the facial expression in

https://www.rd.ntt/cs/event/openhouse/2020/download/c_18_en.pdf

Takuhiro Kaneko | NTT R&D Website

speech) in a controllable manner to make it possible to create or convert media as intended. In this page

https://www.rd.ntt/e/organization/researcher/special/s_046.html

Marc Delcroix | NTT R&D Website

: advances in speech enhancement and recognition for mobile multi-microphone devices」" 2016年 音響学会粟屋 潔学術奨励賞受賞

https://www.rd.ntt/organization/researcher/special/s_033.html

0038.pdf

outstanding problem for speech transmission, enhancement and recognition. The variety and the varying nature

https://www.rd.ntt/cs/team_project/icl/signal/iwaenc03/cdrom/data/0038.pdf

0059.pdf

separated speech, and improved accu- racy of automatic phone recognition on target speech. The preemphasis

https://www.rd.ntt/cs/team_project/icl/signal/ica2003/cdrom/data/0059.pdf

WPE speech dereverberation

speech recognition to prevent the prediction filter from distorting your target signal or amplifying

https://www.rd.ntt/cs/team_project/media/signal/wpe/config.html

メディア情報研究部｜NTTコミュニケーション科学基礎研究所｜NTT R&D Website

. Ochiai, M. Delcroix, "Frontend Token Enhancement for Token-Based Speech Recognition" ・Y. Ren, N. Harada

https://www.rd.ntt/cs/team_project/media/

Thought-processing Technology for Understanding, Reproducing, and Extending Human Thinking Ability｜NTT R&D Website

, speech recognition, and thought-enhancing stimulus design are introduced. Kyosuke Nishida, Takeshi

https://www.rd.ntt/e/research/JN202110_15620.html

Main Topics | NTT Communication Science Laboratories | NTT R&D Website

High fidelity color reproduction and analysis Speech Recognition for Computers Listening to human

https://www.rd.ntt/e/cs/research_topic/

NTT Communication Science Laboratories Open House 2017

precisely using LSTM framework 3. Multilingual Speech Recognition: Support 10 languages in Asia-Pacific

https://www.rd.ntt/cs/event/openhouse/2017/exhibition/19/index_en.html

Recognition Research Group | NTT Communication Science Laboratories | NTT R&D Website

Recognition Research Group | NTT Communication Science Laboratories | NTT R&D Website NTT R&D

https://www.rd.ntt/e/cs/team_project/media/recognition/

スライド 1

Voice command and speech communication in car - World’s best voice capture and recognition technologies

https://www.rd.ntt/cs/event/openhouse/2019/download/15_c_en.pdf

poster_en.pdf

). WFST has been successfully utilized in the speech recognition field, where it is shown that it can

https://www.rd.ntt/cs/event/openhouse/2016/exhibition/16/poster_en.pdf

NTT Communication Science Laboratories Open House 2016

Recognizing your voice even in noisy environments - Advances in distant speech recognition technologies

https://www.rd.ntt/cs/event/openhouse/2016/exhibition/19/index_en.html

メディア情報研究部　過去のニュース｜NTTコミュニケーション科学基礎研究所｜NTT R&D Website

2025/12 【表彰】森谷准特別研究員、小川主任研究員らによる以下の論文が、国際会議 2025 IEEE Automatic Speech Recognition and Understanding

https://www.rd.ntt/cs/team_project/media/past_news.html

Ryo Masumura | NTT R&D Website

training data sets for various media processing including automatic speech recognition and natural language

https://www.rd.ntt/e/organization/researcher/special/s_035.html

NTT Communication Science Laboratories Open House 2019

Voice command and speech communication in car - World’s best voice capture and recognition technologies

https://www.rd.ntt/cs/event/openhouse/2019/exhibition15/index_en.html

NTT Communication Science Laboratories Open House 2014 Program

integration for speech enhancement - Defeat reverberation: enemy of speech recognition - Advanced speech

https://www.rd.ntt/cs/event/openhouse/2014/program_en.html

0056.pdf

-end processing of ASR (Automatic Speech Recognition). AMNOR (Adaptive Microphone-array for NOise

https://www.rd.ntt/cs/team_project/icl/signal/iwaenc03/cdrom/data/0056.pdf

Listening to human speech in noisy reverberant environments | NTT Communication Science Laboratories | NTT R&D Website

automatic speech recognition (ASR) in a real world environment. Although many enhancement methods have

https://www.rd.ntt/e/cs/team_project/media/signal/research_media06.html

NTT Communication Science Laboratories Open House 2018

. For example, in research on speech recognition, we are moving from the situation of a single person

https://www.rd.ntt/cs/event/openhouse/2018/talk/director/index_en.html

poster_en.pdf

between robots to avoid dialogue breakdown. Avoid dialogue breakdown in case of speech recognition error

https://www.rd.ntt/cs/event/openhouse/2016/exhibition/15/poster_en.pdf

NTT Communication Science Laboratories Open House 2019

Learning unknown objects from speech and vision - Crossmodal audio-visual concept discovery - Abstract In

https://www.rd.ntt/cs/event/openhouse/2019/exhibition20/index_en.html

Human Behavior Recognition from Multimodal Information | NTT Communication Science Laboratories | NTT R&D Website

detection from images or videos Speech activity detection from audio signals Facial expression recognition

https://www.rd.ntt/e/cs/team_project/human/resonance/research_human06.html

0037.pdf

performance is illustrated by speech recognition ex- periments on real recordings corrupted by babble noise

https://www.rd.ntt/cs/team_project/icl/signal/ica2003/cdrom/data/0037.pdf

NTT Communication Science Laboratories Open House 2017 Program

speech identification & recognition technology Personalizing your speech recognizer Neural network

https://www.rd.ntt/cs/event/openhouse/2017/program_en.html

Microsoft Word - ica_new.doc

robustness of speech recognition algorithms. However, depending on the type of emotion or task induced stress

https://www.rd.ntt/cs/team_project/icl/signal/ica2003/cdrom/data/0182.pdf

NTT Communication Science Laboratories Open House 2020

in speech through the facial expression in a face image. The proposed system consists of performing

https://www.rd.ntt/cs/event/openhouse/2020/exhibition18/index_en.html

poster_en19.pdf

better at annotating an audio signal than a baseline method. Yasunori Ohishi Media Recognition Research

https://www.rd.ntt/cs/event/openhouse/2013/exhibition/media7/poster_en19.pdf

スライド 1

Learning unknown objects from speech and vision - Crossmodal audio-visual concept discovery - In order for

https://www.rd.ntt/cs/event/openhouse/2019/download/20_c_en.pdf

talk_yoshioka.pdf

, M. Picheny, “The IBM 2015 English conversational telephone speech recognition system,” arXiv:1505

https://www.rd.ntt/cs/event/openhouse/2016/talk/research1/talk_yoshioka.pdf

NTT Communication Science Laboratories Open House 2016 Program

deep learning and signal processing that are making speech recognition leap forward - Takuya Yoshioka

https://www.rd.ntt/cs/event/openhouse/2016/program_en.html

Major issues and research trends in the evolution of artificial intelligence (AI)｜NTT R&D Website

to “listen” and “speak,” NTT is conducting case study research on “speech recognition and spoken

https://www.rd.ntt/e/ai/0001.html

Abstracts of plenary talks

environmental noise. In fact, in noisy environments the performance of speech coders or speech recognition

https://www.rd.ntt/cs/team_project/icl/signal/iwaenc03/plenary_abst.html

NTT Communication Science Laboratories Open House 2018

between the speech recognition and synthesis techniques and the human brain mechanisms will be discussed

https://www.rd.ntt/cs/event/openhouse/2018/talk/research4/index_en.html

NTT Communication Science Laboratories Open House 2014

speech intelligibility and automatic speech recognition performance. Even though reverberation is

https://www.rd.ntt/cs/event/openhouse/2014/talk/research2/index_en.html

NTT コミュニケーション科学基礎研究所 オープンハウス2019

. Watanabe, T. Iwata, A. Ogawa, M. Delcroix, “Semi-supervised end-to-end speech recognition,” in Proc

https://www.rd.ntt/cs/event/openhouse/2019/exhibition16/

スライド 1

speech recognition,” in Proc. Interspeech, 2018. [2] S. Karita, S. Watanabe, T. Iwata, M. Delcroix, A

https://www.rd.ntt/cs/event/openhouse/2019/download/16_c.pdf

NTT Communication Science Laboratories Open House 2016

language lectures by using speech recognition and machine translation technology. The system allows non

https://www.rd.ntt/cs/event/openhouse/2016/exhibition/23/index_en.html

poster_23.pdf

誤りから学ぶ音声認識システム ～音声認識のための識別モデル学習法～ Speech Recognition System that Learns from Errors

https://www.rd.ntt/cs/event/openhouse/2011/exhibition/23/poster_23.pdf

F04_leaf_e.pdf

separated without relying on speech recognition, which tends to be less accurate for signals with multiple

https://www.rd.ntt/forum/2023/doc/F04_leaf_e.pdf

P04_leaf_e.pdf

people Speech recognition technology that can operate in real-time at all the times, even on smart

https://www.rd.ntt/forum/2023/doc/P04_leaf_e.pdf

poster_en.pdf

speech recognition, computer vision and AI. However, Deep Learning is time-consuming for the training. In

https://www.rd.ntt/cs/event/openhouse/2016/exhibition/7/poster_en.pdf

Using Neural Networks to Enable Computers to Listen to Voices Selectively as Humans Do | NTT R&D Website

research was on speech recognition, which involved adapting a speech-recognition system to the voice of a

https://www.rd.ntt/e/research/JN202506_34173.html

Demonstration of speech enhancement in highly non-stationary multisource environment (CHiME Challenge)

., Souden, M., Hahm S.-J. and Nakamura, A., ``Speech Recognition in the Presence of Highly Non-Stationary

https://www.rd.ntt/cs/team_project/media/signal/CHiME_demo/

Communication with Desired Voice | NTT R&D Website

combine automatic speech recognition (ASR) and text-to-speech synthesis. This approach uses ASR to

https://www.rd.ntt/e/research/JN202009_6715.html

NTT Communication Science Laboratories Open House 2018

. Reference [1] A. Ogawa, M. Delcroix, S. Karita, T. Nakatani, “Rescoring n-best speech recognition list based

https://www.rd.ntt/cs/event/openhouse/2018/exhibition/17/index_en.html

0078.pdf

SPEECH ENHANCEMENT AND RECOGNITION IN CAR ENVIRONMENT USING BLIND SOURCE SEPARATION AND SUBBAND

https://www.rd.ntt/cs/team_project/icl/signal/ica2003/cdrom/data/0078.pdf

Publications | NTT Communication Science Laboratories | NTT R&D Website

(Japanese only). ｢Speech Recognition Algorithms Using Weighted Finite-State Transducers｣ WrittenTakaaki Hori

https://www.rd.ntt/e/cs/overview/publications.html

金子 卓弘 | NTT R&D Website

on Pattern Recognition (ICPR 2012) 2017 2016年度 音声研究会 研究奨励賞 2020 令和元年度 東京大学大学院 情報理工学系研究科 研究科長賞 社内表彰

https://www.rd.ntt/organization/researcher/special/s_046.html

NTT Communication Science Laboratories Open House 2013 Program

- Research on multi-speaker conversational speech recognition: past, present and future - Takaaki Hori, Media

https://www.rd.ntt/cs/event/openhouse/2013/program_en.html

poster18.pdf

speech recognition based on WFST structured classifiers and deep bottleneck features,” in Proc. The 38th

https://www.rd.ntt/cs/event/openhouse/2013/exhibition/media6/poster18.pdf

Torch Relay Regional Event × Voice-recognition Communication Technology｜NTT R&D Website

Torch Relay Regional Event × Voice-recognition Communication Technology｜NTT R&D Website NTT R&D

https://www.rd.ntt/e/research/JN202111_16140.html

メディア認識研究グループ｜NTTコミュニケーション科学基礎研究所｜NTT R&D Website

in Human Action Recognition. Pattern Recognition, 146. 国際会議予稿 Yu Mitsuzumi, Akisato Kimura, Go Irie

https://www.rd.ntt/cs/team_project/media/recognition/

スライド 1

, M. Delcroix, S. Karita, T. Nakatani, “Rescoring n-best speech recognition list based on one-on-one

https://www.rd.ntt/cs/event/openhouse/2018/exhibition/17/poster_en_17.pdf

NTT Communication Science Laboratories Open House 2019

recognition technologies Learning Speech Recognition from Small Paired Data Semi-supervised end-to-end

https://www.rd.ntt/cs/event/openhouse/2019/program_en.html

iwaenc2003-3.dvi

usually contain reverberation, which degrades speech recognition performance. This is especially severe

https://www.rd.ntt/cs/team_project/icl/signal/iwaenc03/cdrom/data/0049.pdf

A Close Link Between "Hearing" and "Speaking" | NTT Communication Science Laboratories | NTT R&D Website

signals. Future Implications As automatic speech recognition technologies advance, the challenges are

https://www.rd.ntt/e/cs/team_project/human/smr/research_human02.html

複数人会話理解｜NTTコミュニケーション科学基礎研究所｜NTT R&D Website

Emotion Recognition Based on Large-Scale Automatic Speech Recognizer," Proc. ICASSP, 2025 R. Fukuda et al

https://www.rd.ntt/cs/team_project/media/signal/research_signal03.html

NTT Communication Science Laboratories Open House 2014 Making computers listen to desired sounds anywhere - Probabilistic modeling and integration for speech enhancement -

and reverberation, which degrades both sound quality and speech recognition accuracy. Conventional

https://www.rd.ntt/cs/event/openhouse/2014/exhibition/19/index_en.html

スライド 1

to EUSIPCO2018. [2] K. Tanaka, T. Kaneko, N. Hojo, H. Kameoka, “Synthetic-to-natural speech

https://www.rd.ntt/cs/event/openhouse/2018/exhibition/20/poster_en_20.pdf

Researching AI Technologies to Support and Substitute Human Activities｜NTT R&D Website

to achieve the full potential of AI. Taking speech-recognition technology as an example, differences

https://www.rd.ntt/e/research/JN202010_7180.html

Expressive Speech-synthesis Technology Supports People's Daily Lives | NTT R&D Website

researching speech processing ever since my student days. At that time, I was researching speech recognition

https://www.rd.ntt/e/research/JN202504_33343.html

poster_en_17.pdf

interested in developing a pronunciation conversion system that can convert non-native speech into

https://www.rd.ntt/cs/event/openhouse/2017/exhibition/17/poster_en_17.pdf

D01-21-e.pdf

recognition, and speech synthesis, it has become possible for digital humans to respond flexibly according to

https://www.rd.ntt/forum/2024/doc/D01-21-e.pdf