Multi-objective Approach to Speech Enhancement Using Tunable Q-Factor-based Wavelet Transform and ANN Techniques

Dash, Tusar Kanti; Solanki, Sandeep Singh; Panda, Ganapati

doi:10.1007/s00034-021-01753-2

Cited by 9 publications

(4 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning-based (ML) classifiers working along with time and frequency extracted features have made substantial progress in this field. Even in noisy conditions, this combination exhibited outstanding accuracies for discrete sound categorization (Dash et al, 2021b ). To initiate classification, all the above-mentioned 78 features were extracted from the speech signal and were provided as inputs to the following classifiers.…”

Section: Methodsmentioning

confidence: 99%

“…Neural network-based classifier models are widely used in speech processing for improved performance (Lopez-Moreno et al, 2016;Dash et al, 2020). In this case, feed-forward fully connected neural network (FCNN) is used with the input layer connected to a fully connected layer of 10 neurons, a ReLU function, followed by a second fully connected layer, a softmax function.…”

Section: Feed-forward Fully Connected Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Speech phoneme and spectral smearing based non-invasive COVID-19 detection

Mishra

Dash²,

Panda³

2023

Front. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

COVID-19 is a deadly viral infection that mainly affects the nasopharyngeal and oropharyngeal cavities before the lung in the human body. Early detection followed by immediate treatment can potentially reduce lung invasion and decrease fatality. Recently, several COVID-19 detections methods have been proposed using cough and breath sounds. However, very little study has been done on the use of phoneme analysis and the smearing of the audio signal in COVID-19 detection. In this paper, this problem has been addressed and the classification of speech samples has been carried out in COVID-19-positive and healthy audio samples. Additionally, the grouping of the phonemes based on reference classification accuracies have been proposed for effectiveness and faster detection of the disease at a primary stage. The Mel and Gammatone Cepstral coefficients and their derivatives are used as the features for five standard machine learning-based classifiers. It is observed that the generalized additive model provides the highest accuracy of 97.22% for the phoneme grouping “/t//r//n//g//l/.” This smearing-based phoneme classification technique can also be used in the future to classify other speech-related disease detections.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Feed-forward Fully Connected Neural Networkmentioning

confidence: 99%

Speech phoneme and spectral smearing based non-invasive COVID-19 detection

Mishra

Dash²,

Panda³

2023

Front. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

show abstract

“…FLANN is one of the computationally efficient and effective forms of the traditional neural network model [35,36]. Recently, the PFLANN model has been developed for intelligent water fountain sound pleasantness monitoring [37].…”

Section: B Pflann Modelmentioning

confidence: 99%

Cascaded PFLANN Model for Intelligent Health Informatics in Detection of Respiratory Diseases from Speech Using Bio-inspired Computation

Jagannath Dayal Pradhan,

L V Narasimha Prasad,

Tusar Kanti Dash

et al. 2024

JAIT

View full text Add to dashboard Cite

Due to the recent developments in communications technology, cognitive computations have been used in smart healthcare techniques that can combine massive medical data, Artificial intelligence, federated learning, Bio-inspired computation, and the Internet of Medical Things. It has helped in knowledge sharing and scaling ability between patients, doctors, and clinics for effective treatment of patients. Speech-based respiratory disease detection and monitoring are crucial in this direction and have shown several promising results. Since the subject’s speech can be remotely recorded and submitted for further examination, it offers a quick, economical, dependable, and non-invasive prospective alternative detection approach. However, the two main requirements of this are higher accuracy and lower computational complexity and, in many cases, these two requirements do not correlate with each other. This problem has been taken up in this paper to develop a low computational complexity-based neural network with higher accuracy. A Cascaded Perceptual functional link artificial neural network (PFLANN) is used to capture the non-linearity in the data for better classification performance with low computational complexity. The proposed model is being tested for multiple respiratory diseases and the analysis of various performance matrices demonstrates the superior performance of the proposed model both in terms of accuracy and complexity.

show abstract

“…It provides efficient prediction performance when the data size is small with less number of features [21]. FLANN model has been used successfully for the prediction of speech quality, speech enhancement, and other speech-related information [22]. In [23], three neural network models such as FLANN, polynomial perceptron network, and Legendre neural network have been used to predict the machinery noise in opencast mines for a case study in the coal mine of Orissa, India.…”

Section: Introductionmentioning

confidence: 99%

Design of P-FLANN Model for Intelligent Water Fountain Sound Pleasantness Monitoring Using Bio-inspired Computing and Human Speech Perception

Brahma¹,

Dash²,

Panda³

et al. 2023

JAIT

View full text Add to dashboard Cite

Cognitive-inspired Computational Computing systems play a crucial role in designing intelligent health monitoring systems which help both patients and hospitals. It also helps in early and consistent decision-making for various health issues including human psychological health. Water fountains built in parks and public spaces are used as decorative instruments which not only give appealing visuals but also it provides a relaxing environment to the visitors. These natural sounds have a direct effect on the psychological health of visitors. Very few research works are reported on developing the relationship between water sounds and their corresponding psychological impact. This assessment needs trained manpower and a lot of experimental time which is costly and may not be always available. In this paper to access the human health-friendly water fountain sounds from the pleasantness, a Perceptually Weighted functional link artificial neural network (P-FLANN) model is developed. To reduce the computational complexity of training and for faster convergence, swam intelligence-based optimization algorithm is used for updating the weights. It is observed from the comparative simulation results that the proposed P-FLANN model can effectively perform prediction tasks which is not only cost-effective but also 95% accurate and can play a crucial role in designing human health-friendly water fountains in smart cities.

show abstract

Multi-objective Approach to Speech Enhancement Using Tunable Q-Factor-based Wavelet Transform and ANN Techniques

Cited by 9 publications

References 60 publications

Speech phoneme and spectral smearing based non-invasive COVID-19 detection

Speech phoneme and spectral smearing based non-invasive COVID-19 detection

Cascaded PFLANN Model for Intelligent Health Informatics in Detection of Respiratory Diseases from Speech Using Bio-inspired Computation

Design of P-FLANN Model for Intelligent Water Fountain Sound Pleasantness Monitoring Using Bio-inspired Computing and Human Speech Perception

Contact Info

Product

Resources

About