ATBECHDEDKESFRGBGRITLILUNLSEMCPTIESILTLVFIROMKCYALTRBGCZEEHUPLSK..HRIS..MTNORS..SM..................*J2.0.21 - 2999001/03698245CORRECTED EUROPEAN PATENT SPECIFICATIONB8B120230809W173deBibliographieenBibliographyfrBibliographieEP17795120.92017102020200318enenen2023080920233220200826202035202306282023262023011620230809202332G06N 3/04 20230101AFI20221219BHEP G06N 3/063 20230101ALI20221219BHEP G06F 9/50 20060101ALI20221219BHEP G06F 9/5066 20130101 FI20180702BHEP G06N 3/063 20130101 LI20180702BHEP G06N 3/044 20230101 LI20230101RHEP dePARALLELE VERARBEITUNG FÜR SIGNALERZEUGENDE NEURONALE NETZEenPARALLEL PROCESSING FOR SIGNAL GENERATION NEURAL NETWORKSfrTRAITEMENT PARALLÈLE POUR RÉSEAUX NEURONAUX DE GÉNÉRATION DE SIGNAUXSercan O Arik ET AL: "Deep Voice: Real-time Neural Text-to-Speech", , 24 February 2017 (2017-02-24), XP055489867, Retrieved from the Internet: URL:https://arxiv.org/pdf/1702.07825.pdfMasoud Abdi ET AL: "Multi-Residual Networks: Improving the Speed and Accuracy of Residual Networks", , 19 September 2016 (2016-09-19), XP055489868, Retrieved from the Internet: URL:https://arxiv.org/pdf/1609.05672.pdfJAFRI SYED MOHAMMAD ASAD HASSAN ET AL: "SPEED: Open-Source Framework to Accelerate Speech Recognition on Embedded GPUs", 2017 EUROMICRO CONFERENCE ON DIGITAL SYSTEM DESIGN (DSD), IEEE, 30 August 2017 (2017-08-30), pages 94-101, XP033157617, DOI: 10.1109/DSD.2017.89 [retrieved on 2017-09-25]ELSEN, Erich Konrad1600 Amphitheatre ParkwayMountain View, California 94043USDIELEMAN, Sander Entienne Lea1600 Amphitheatre ParkwayMountain View, California 94043USDeepMind Technologies Limited101822405PC928342EP5 New Street SquareLondon Greater London EC4A 3TWGBMartin, Philip John100046500Marks & Clerk LLP 62-68 Hills RoadCambridge CB2 1LAGBALATBEBGCHCYCZDEDKEEESFIFRGBGRHRHUIEISITLILTLULVMCMKMTNLNOPLPTRORSSESISKSMTRUS201705766320171020enWO201907889020190425201917