Results 61 to 70 of about 7,229,432 (222)

Fault Diagnosis of Wind Turbine Gearbox Based on Mel Spectrogram and Improved ResNeXt50 Model

Applied Sciences
In response to the problem of complex and variable loads on wind turbine gearbox bearing in working conditions, as well as the limited amount of sound data making fault identification difficult, this study focuses on sound signals and proposes an ...
Xiaojuan Zhang, Feixiang Jia, Yayu Chen
semanticscholar   +1 more source

Intelligent Maintenance Review for Robots: Multimodal Information, Deep Diagnosis and Embodied Artificial Intelligence

Advanced Robotics Research, EarlyView.
This review maps the methods to monitor robots’ health by fusing vibration, sound, control signals, vision, force, and oil information with artificial intelligence. It identifies deep learning, transfer learning, digital twins, and physics‐informed models as key methodological pathways enabling earlier diagnosis, safer human–robot collaboration, and ...
Yuting Qiao, Hongbo Wang, Naipeng Li, Shuo Wang, Yaguo Lei, Tonghai Wu, Junyi Cao   +6 more
wiley   +1 more source

A Unified Approach to Voice Classification: Leveraging Spectrograms, Mel Spectrograms, and Statistical Features

IEEE Access
This study presents a multi-input neural network architecture for voice classification that integrates two parallel convolutional neural networks (CNNs) for spectrogram and Mel spectrogram images, along with a fully connected dense network for six handpicked numerical statistical features from time domain signal.
Muhammad Talha, Huma Ghafoor, Seung Yeob Nam   +2 more
openaire   +2 more sources

Resource‐Aware Contrastive Scattering Meta‐Learning for Efficient Few‐Shot Acoustic Anomaly Detection

Advanced Intelligent Systems, EarlyView.
This paper introduces a resource‐aware Contrastive Scattering Meta‐Learning (CSML) framework for acoustic anomaly detection. By leveraging training‐free wavelet scattering and metric‐based meta‐learning, the model achieves competitive performance with only 50 K learnable parameters—a 98% reduction compared to state‐of‐the‐art frameworks—enabling ...
Rami Zewail, Bassem Mokhtar
wiley   +1 more source

Mel-spectrogram augmentation for sequence to sequence voice conversion

CoRR, 2020
For training the sequence-to-sequence voice conversion model, we need to handle an issue of insufficient data about the number of speech pairs which consist of the same utterance. This study experimentally investigated the effects of Mel-spectrogram augmentation on training the sequence-to-sequence voice conversion (VC) model from scratch.
Hwang, Yeongtae, Cho, Hyemin, Yang, Hongsun, Won, Dong-Ok, Oh, Insoo, Lee, Seong-Whan   +5 more
openaire   +3 more sources

Passive Acoustic Identification of Social Groups in the Hainan Gibbon

Remote Sensing in Ecology and Conservation, EarlyView.
Passive acoustic monitoring offers a non‐invasive means of assessing visually hard‐to‐survey wildlife species with distinctive vocalizations. We evaluated whether deep learning can identify Hainan gibbon (Nomascus hainanus) social groups from their calls.
Emmanuel Kabuga, Emmanuel Dufourq, Nonhlanhla L. Luphade, Samuel T. Turvey, Heidi Ma, Susan M. Cheyne, Hui Liu, Jessica V. Bryant, Qing Chen, Wenyong Li, Zhiwei Liu, Zhaoli Zhou, Stefan Britz, Bubacarr Bah, Ian Durbach   +14 more
wiley   +1 more source

Artificial Intelligence in Voice Disorders: Current Landscape, Emerging Applications and Future Directions

World Journal of Otorhinolaryngology - Head and Neck Surgery, EarlyView.
ABSTRACT Objective To provide a comprehensive review of the current landscape of artificial intelligence (AI) applications in voice disorder, with emphasis on emerging applications, limitations, and future directions for clinical integration. Methods Literature review.
Rachel B. Kutler, Anaïs Rameau
wiley   +1 more source

Is GAN Necessary for Mel-Spectrogram-Based Neural Vocoder?

IEEE Signal Processing Letters
Accepted by IEEE Signal Processing ...
Hui-Peng Du, Yang Ai, Rui-Chen Zheng, Ye-Xin Lu, Zhen-Hua Ling   +4 more
openaire   +2 more sources

GELP: GAN-Excited Liner Prediction for Speech Synthesis from Mel-Spectrogram [PDF]

, 2019
Recent advances in neural network -based text-to-speech have reached human level naturalness in synthetic speech. The present sequence-to-sequence models can directly map text to mel-spectrogram acoustic features, which are convenient for modeling, but ...
Bajibabu Bollepalli, Paavo Alku, Bollepalli, Bajibabu, Alku, Paavo, Lauri Juvela, Junichi Yamagishi, Yamagishi, Junichi, Juvela, Lauri   +7 more
core   +1 more source

DrLS: Distortion‐Resistant Lossless Steganography via Colour Depth Interpolation

CAAI Transactions on Intelligence Technology, EarlyView.
ABSTRACT The lossless data steganography is to hide a certain amount of information into a container image. Previous lossless steganography methods fail to strike a balance between capacity, imperceptibility, accuracy, and robustness, commonly vulnerable to distortion on container images.
Youmin Xu, Xuanyu Zhang, Xiandong Meng, Jian Zhang   +3 more
wiley   +1 more source