2020-2022 YAMAGATA UNIVERSITY Research Seeds Collection
42/90

40Convert to a voice that is easy to hearIllustrationMealInformationCollectionandAdviseSystem.NetworkProtocol.SecuritySystem.VoiceconversionsystemSpeechenhancementsysteminrealContent:InformationandcommunicationnetworksarethesocialContentinfrastructurethatcreatesnewvalueinallsocial,economicandculturalactivitiesofthe21stcentury.Inthislaboratory,wewill Information and communication networks are the social infrastructure that creates new value in all social, economic and developtechnologiesthatwillcontributetotheconstructionofbasiccultural activities of the 21st century. In this laboratory, we will technologiesforthenext-generationInternet,theInternetofThingsdevelop technologies that will contribute to the construction of (IoT)andvariouswirelesscommunications,anddevelopsystemsbasic technologies for the next-generation Internet, the Internet of Things (IoT) and various wireless communications, and develop thatareusefullifeusingtheInternetandIoT.Wearesystems that are useful in real life using the Internet and IoT. We conductingresearchanddevelopmentonfourmajorthemes:networkare conducting research and development on four major themes: protocols,medical/healthcarerelatedsystems,IoTrelatedsystems,network protocols, medical/healthcare related systems, IoT related systems, and security related systems. andsecurityrelatedsystems.. In network protocol research, we are conducting research on Innetworkprotocolresearch,weareconductingresearchonprotocols applicable to IoT, wireless mesh networks, delay protocolsapplicabletoIoT,wirelessmeshnetworks,delaytoleranttolerant networks (DTN), sensor networks, and ad hoc networks networks(DTN),sensornetworks,andadhocnetworksusingusing machine learning and heuristics method. Regarding network systems, research and development of medical diagnosis support machinelearningandheuristicsmethod.Regardingnetworksystems,systems that support doctors' diagnosis, health advice systems researchanddevelopmentofmedicaldiagnosissupportsystemsthatthat use vital data, and systems that automatically collect meal supportdoctors'diagnosis,healthadvicesystemsthatusevitaldata,information and provide advice are conducted through machine learning and fuzzy theory. andsystemsthatautomaticallycollectmealinformationandprovideadviceareconductedthroughmachinelearningandfuzzytheory.Special objectivesAppealingpoint:Wewillactivelyengageinindustry-academiacollaboration.WewillcontributetothedevelopmentofIoTandAI We will actively engage in industry-academia collaboration. We byutilizingourresearchcontent.will contribute to the development of IoT and AI by utilizing our research content.Yamagata UniversityGraduate School of Science and Engineering Yamagata University Graduate School of Science and Engineering Research Interest : Research InterestInformation Network EngineeringE-mail :akoyama@yz.yamagata-u.ac.jpTel :+81-238-26-3577E-mail ・ akoyama@yz.yamagata-u.ac.jpTel ・ +81-238-26-3577HP :https://koyamalab.yz.yamagata-u.ac.jp/HP・https://koyamalab.yz.yamagata-u.ac.jp/InformationNetworkEngineeringContentContent::Speechprocessingincludesawiderangeoftechnologies,suchas Speech processing includes a wide range of technologies, speechrecognition,speechsynthesis,andvoiceconversion.Oursuch as speech recognition, speech synthesis, and voice laboratoryisengagedinvariousresearchworkonspeechprocessing.conversion. Our laboratory is engaged in various research Machinelearningisthebasisofspeechtechnology.Deeplearning,work on speech processing.whichhasrecentlybecomeahottopic,isalsoatypeofmachine Machine learning is the basis of speech technology. Deep learning.Weareworkingondevelopingvariousspeechtechnologieslearning, which has recently become a hot topic, is also a usingcomputerswithGPUscapableofhigh-speedcomputation.type of machine learning. We are working on developing Basedontheresultsofindustry-academiacollaboration,wevarious speech technologies using computers with GPUs successfullydevelopedanunderwatertransceiverforleisuredivers.capable of high-speed computation. Thisdevicecanenabledirectconversationunderwater.Presently, Based on the results of industry-academia collaboration, thedeviceisusednotonlyforleisure,butalsoforwaterrescue.we successfully developed an underwater transceiver for leisure divers. This device can enable direct conversation Appealing point::under water. Presently, the device is used not only for Wearedevelopingtechniquesforspeechandaudioprocessingleisure, but also for water rescue.usingmachinelearningmethodssuchasdeeplearning.Regardingindustry-academiacollaboration,wearecurrentlyworkingonproductinspectionbysound. We are developing techniques for speech and audio processing using machine learning methods such as deep learning. Regarding industry-academia collaboration, we are currently working on product inspection by sound.Yamagata UniversityGraduate School of Science and Engineering Research Interest :Speech and Audio ProcessingYamagata University Graduate School of Science and EngineeringE-mail :tkosaka[at]yz.yamagata-u.ac.jpResearch InterestTel :+81-238-26-3369Speech and Audio ProcessingFax:+81-238-26-3369E-mail ・ tkosaka[at]yz.yamagata-u.ac.jpTel ・ +81-238-26-3369HP :https://speech-lab.yz.yamagata-u.ac.jp/Fax ・ +81-238-26-3369HP・https://speech-lab.yz.yamagata-u.ac.jp/pSpecial objectivesSpeech and Audio Processing using Machine LearningSpeech and AudioProcessing using Machine LearningVoiceconversionSpeechenhancementProfessor Tetsuo KosakaProfessor Tetsuo Kosaka   Professor Akio KoyamaProfessor Akio KoyamaResearch and Development of NetworkProtocolandNetworkSystemResearch and Development of Network Protocol and Network SystemVoice of Mr. AVoice of Mrs. B

元のページ  ../index.html#42

このブックを見る