Results 101 to 110 of about 3,717 (149)
PESQ and STOI result for the VoiceBank dataset.
Long short-term memory (LSTM) has been effectively used to represent sequential data in recent years. However, LSTM still struggles with capturing the long-term temporal dependencies.
Amil Daraz (9667034) +3 more
core +1 more source
Employing Huber and TAP Losses to Improve Inter-SubNet in Speech Enhancement
In this study, improvements are made to Inter-SubNet, a state-of-the-art speech enhancement method. Inter-SubNet is a single-channel speech enhancement framework that enhances the sub-band spectral model by integrating global spectral information, such ...
Jeih-Weih Hung +2 more
doaj +1 more source
Du contrôle au conseil : 40 ans de CAUE. Témoignage de Henri Le Pesq
Du 27 février au 3 avril 2017, Politiques de la culture met en ligne les communications et les témoignages donnés à l'occasion de la journée d'étude sur le 40e anniversaire de la loi sur l'architecture Témoignage d'Heni Le Pesq, ancien directeur du CAUE
chmc1
core
Multimode Tree Coding of Speech with Backward Pitch Prediction and Perceptual Pre-and Post-weighting
-A low delay and low complexity Multimode Tree Coder with backward pitch predictor is proposed. For the Multimode Tree Coding, the speech is classified into five different modes, and each mode is coded at a suitable bit-rate using a tree coder with ...
Ying-Yi Li, Jerry D Gibson
core
Audio-visual speech enhancement with multi-level feature deep fusion under low signal-to-noise ratio
To address the limitations in feature extraction and cross-modal fusion in audio-visual speech enhancement, a multistage deep fusion method was proposed for low signal-to-noise ratio (SNR) conditions.
ZHANG Tianqi +3 more
doaj
Mechanical properties optimization in low carbon steel by microstructure modification
This paper discusses the formation of microstructures with different volume fractions, as an outcome of a specific heat treatment, with the following phases: ferrite, martensite, bainite and retained austenite.
Monteiro, Waldemar Alfredo +3 more
core
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)fepisa - fund. de ensino, pesq.
Uchôa, Thales Brandão [UNESP]
core
Speech quality investigation using PESQ in a simulated climax system for ATM
Validerat; 20101217 (root)
openaire +1 more source
A speech enhancement method based on depthwise separable convolution and MambaUNet
Aiming to enhance the robustness and modeling efficiency of speech enhancement technology in complex noisy environments, a novel speech enhancement network, DW-MambaUNet was proposed, which integrated depthwise separable convolution and a structured ...
MENG Xiangcai +3 more
doaj +2 more sources
Crotoxin-induced behavioral effects in rats
Crotoxin is the major component of Crotalus durissus terrificus venom. In view of the presence of high-affinity specific binding sites for crotoxin in the brain, the objective of this work was to investigate whether crotoxin induces behavioral effects in
Moreira, E. G. [UNESP] +4 more
core

