Results 41 to 50 of about 3,240 (204)

cMelGAN: An Efficient Conditional Generative Model Based on Mel Spectrograms

CoRR, 2022
Analysing music in the field of machine learning is a very difficult problem with numerous constraints to consider. The nature of audio data, with its very high dimensionality and widely varying scales of structure, is one of the primary reasons why it is so difficult to model.
Tracy Qian, Jackson Kaunismaa, Tony Chung   +2 more
openaire   +2 more sources

Data-Dependent Feature Extraction Method Based on Non-Negative Matrix Factorization for Weakly Supervised Domestic Sound Event Detection

Applied Sciences, 2021
In this paper, feature extraction methods are developed based on the non-negative matrix factorization (NMF) algorithm to be applied in weakly supervised sound event detection. Recently, the development of various features and systems have been attempted
Seokjin Lee, Minhan Kim, Seunghyeon Shin, Sooyoung Park, Youngho Jeong   +4 more
doaj   +1 more source

Pre-Configured Deep Convolutional Neural Networks with Various Time-Frequency Representations for Biometrics from ECG Signals

Applied Sciences, 2019
We evaluated electrocardiogram (ECG) biometrics using pre-configured models of convolutional neural networks (CNNs) with various time-frequency representations.
Yeong-Hyeon Byeon, Keun-Chang Kwak
doaj   +1 more source

Mel frequency spectral domain defenses against adversarial attacks on speech recognition systems [PDF]

JASA Express Letters, 2023
Automatic speech recognition (ASR) systems are vulnerable to adversarial attacks due to their reliance on machine learning models. Many of the defenses explored for defending ASR systems simply adapt defense approaches developed for the image domain ...
Nicholas Mehlman, Anirudh Sreeram, Raghuveer Peri, Shrikanth Narayanan   +3 more
doaj   +1 more source

Feature selection for emotion recognition in speech: a comparative study of filter and wrapper methods [PDF]

PeerJ Computer Science
Feature selection is essential for enhancing the performance and reducing the complexity of speech emotion recognition models. This article evaluates various feature selection methods, including correlation-based (CB), mutual information (MI), and ...
Alaa Altheneyan, Aseel Alhadlaq
doaj   +2 more sources

Text-only domain adaptation for end-to-end ASR using integrated text-to-mel-spectrogram generator

, 2023
We propose an end-to-end Automatic Speech Recognition (ASR) system that can be trained on transcribed speech data, text-only data, or a mixture of both. The proposed model uses an integrated auxiliary block for text-based training.
Lavrukhin, Vitaly, Ginsburg, Boris, Bataev, Vladimir, Korostik, Roman, Shabalin, Evgeny   +4 more
core   +2 more sources

A Hybrid CNN and RNN Variant Model for Music Classification

Applied Sciences, 2023
Music genre classification has a significant role in information retrieval for the organization of growing collections of music. It is challenging to classify music with reliable accuracy.
Mohsin Ashraf, Fazeel Abid, Ikram Ud Din, Jawad Rasheed, Mirsat Yesiltepe, Sook Fern Yeo, Merve T. Ersoy   +6 more
doaj   +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

Environment Sound Classification Based on Visual Multi-Feature Fusion and GRU-AWS

IEEE Access, 2020
There are two major questions regarding Environmental Sound Classification (ESC). What is the best audio recognition framework, and what is the most robust audio feature?
Ning Peng, Aibin Chen, Guoxiong Zhou, Wenjie Chen, Wenzhuo Zhang, Jing Liu, Fubo Ding   +6 more
doaj   +1 more source

Empirical Mode Decomposition-Based Feature Extraction for Environmental Sound Classification

Sensors, 2022
In environment sound classification, log Mel band energies (MBEs) are considered as the most successful and commonly used features for classification. The underlying algorithm, fast Fourier transform (FFT), is valid under certain restrictions.
Ammar Ahmed, Youssef Serrestou, Kosai Raoof, Jean-François Diouris   +3 more
doaj   +1 more source