Results 71 to 80 of about 981,169 (268)
Federated Learning for Privacy Preserving On-Device Speaker Recognition
, 2021 State-of-the-art speaker recognition systems are usually trained on a single computer using speech data collected from multiple users. However, these speech samples may contain private information which users are not willing to share.
Zewoudie, Abraham +3 more
core +1 more source
Prosodic Prominence and Boundaries in Sequence-to-Sequence Speech Synthesis [PDF]
, 2020 Recent advances in deep learning methods have elevated synthetic speech quality to human level, and the field is now moving towards addressing prosodic variation in synthetic speech.Despite successes in this effort, the state-of-the-art systems fall ...
Juraj Šimko +7 more
core +1 more source
Advanced Materials, EarlyView.Intrinsic material dynamics are harnessed as computational resources for neuromorphic in‐materio physical reservoir computing. Defects, ionic motion, interfaces, percolation, geometry, and biasing shape transient states that provide fading memory, nonlinearity, and high‐dimensional projection for simple readout. A descriptor‐to‐dynamics framework links
Kshitij RB Singh +5 more
wiley +1 more source
Sensors, 2019 Deep neural networks (DNNs) have been shown to be effective for single sound source localization in shallow water environments. However, multiple source localization is a more challenging task because of the interactions among multiple acoustic signals ...
Zhaoqiong Huang +4 more
doaj +1 more source
Advanced Materials Technologies, EarlyView.Hierarchical multi‐material TPMS lattices are engineered as flexible tactile sensors by combining soft and stiff elastomeric layers with a conformal conductive coating. The bilayer architecture delivers sensitivity at low pressures while maintaining a broad detectable range under large loads, enabling reliable pressure and vibration monitoring for ...
Reza Noroozi +3 more
wiley +1 more source
Avoiding distortions due to speech coding and transmission errors [PDF]
, 1999 We have extended our previous research on a new approach to automatic speech recognition (ASR) in the GSM environment. Instead of recognizing from the decoded speech signal, our system works from the digital speech representation used by the GSM encoder.
Valverde Albacete, Francisco José +8 more
core +1 more source
Advanced Materials Technologies, EarlyView.Dense tactile streams from across the humanoid body converge on collide in a central wiring and data bottleneck. By relocating computation closer to and then into the skin itself, near‐ and in‐sensor architectures, together with neuromorphic computing, chart a path toward perception‐native electronic skin, in which the conversion of stimulus into ...
Mijin Kim +6 more
wiley +1 more source
Autoregressive Modelling for Linear Prediction of Ultrasonic Speech [PDF]
, 2010 Ultrasonic speech is a novel research area with significant applications: as a speech-aid prosthesis for patients with voice box difficulties, silent speech interfaces, secure mode of communication in mobile phones and as a communication medium in high ...
McLoughlin, Ian V. +3 more
core
Speech intelligibility under realistic classroom acoustics
, 2017 Speech recognition is fundamental in everyday communication environments, especially in school classrooms where the teaching-learning process takes place.
Puglisi, Giuseppina Emma +7 more
core +1 more source
Deep Dynamic Network Embedding for Link Prediction
IEEE Access, 2018 Network embedding task aims at learning low-dimension latent representations of vertices while preserving the structure of a network simultaneously. Most existing network embedding methods mainly focus on static networks, which extract and condense the ...
Taisong Li +4 more
doaj +1 more source