Improving dysarthric speech recognition using empirical mode decomposition and convolutional neural network

Yakoub, Mohammed Sidi; Selouani, Sid‐Ahmed; Zaidi, Brahim-Fares; Bouchair, Asma

doi:10.1186/s13636-019-0169-5

Cited by 38 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Compared to the standard system this new approach shows satisfactory results. [3] Arjun et al, (2019), devised a method which is used to correct the stutters found in a speech signal. To avoid the recurrence of same word, the speech is sampled into individual words by using appropriate thresholding and speech energy techniques.…”

Section: Literature Surveymentioning

confidence: 99%

Stuttered Speech Recognition And Classification Using Enhanced Kamnan Filter And Neural Network

Vaidianathan¹,

Subramanian²,

Karthik³

2021

Proceedings of the First International Conference on Computing, Communication and Control System, I3CAC 2021, 7-8 June 2021, Bh

View full text Add to dashboard Cite

Stuttering or stammering assessment is one of the vital factors in speech recognition algorithms. To reconstruct the stuttered speech into spontaneous speech it is necessary to detect and correct the features influencing the speech signal. In this paper the speech signal is processed based on the disturbances created by acoustic effects like pauses and noises made both externally and internally. To eliminate the effects of noise on speech signal an Enhanced Kalman Filter is introduced here and its performance along with various filters are studied and compared based on the parameters like Mean Square Error (MSE), Mean Absolute Error (MAE), SNR ratio, Peak Signal to Noise ratio and Cross correlation. Then based on the extracted features classification of the speech signal is carried out using Convolutional Neural Network (CNN) algorithm of Deep learning technique.

show abstract

Section: Literature Surveymentioning

confidence: 99%

Stuttered Speech Recognition And Classification Using Enhanced Kamnan Filter And Neural Network

Vaidianathan¹,

Subramanian²,

Karthik³

2021

Proceedings of the First International Conference on Computing, Communication and Control System, I3CAC 2021, 7-8 June 2021, Bh

View full text Add to dashboard Cite

show abstract

“…According to reference [8], the pre-processing stage utilizes the speech improvement strategy of EMDH to enhance the speech quality of dysarthria. From the EMDH-processed speech, the cepstral coefficients of Mel frequency are extracted and sent into a CNN-based recognizer as input characteristics.The findings imply that the CNN is capable of retrieving latent characteristics of dysarthria speech and that it may be trained faster with fewer data.…”

Section: Related Workmentioning

confidence: 99%

Leveraging Classification of Brain Tumour using Deep Learning Architectures

Karunya¹,

G.R²,

Swathi³

2021

Proceedings of the First International Conference on Combinatorial and Optimization, ICCAP 2021, December 7-8 2021, Chennai, In

View full text Add to dashboard Cite

Despite advances in human intellect and biomedical in the last few decades, people continue to suffer from various cancers due to their volatile nature. This disease is still a major issue for the entire humanity. Brain tumour is one of the most crucial and serious illnesses. Oftheentire primary central nervous system tumours, Brain tumorsmake up 85 to 90%. It isestimatedthat 18,600 adults, including 8,100 women and 10,500 men, will die of primary cancerous tumors of the brain and centralnervoussystem tumors this year. Among the children of various age groups also it is seen as one of the most crucial cancers. Thus, accurate and timely handling of this disease is decisive. In order to speed up the process of brain tumour detection (augmented with accuracy, reliability and experience)Deep learning models can be used.To efficiently diagnose brain tumour kinds and compare classification performance, the proposed work makes optimal use of a newly modelled Convolutional Neural Network and ResNet 50,a pre-trained network.

show abstract

“…The experiment results of this study showed that the CNN-based feature extraction from the MFCC map provided better word-recognition results than other conventional feature extraction methods. More recently, Yakoub et al [43] proposed an empirical model decomposition and Hurst-based model selection (EMDH)-CNN system to improve the recognition of dysarthric speech. The results showed that the proposed system provided higher accuracy than the hidden Markov with Gaussian Mixture model and the CNN model by 20.72% and 9.95%, respectively.…”

Section: Introductionmentioning

confidence: 99%

A Speech Command Control-Based Recognition System for Dysarthric Patients Based on Deep Learning Technology

et al. 2021

View full text Add to dashboard Cite

Voice control is an important way of controlling mobile devices; however, using it remains a challenge for dysarthric patients. Currently, there are many approaches, such as automatic speech recognition (ASR) systems, being used to help dysarthric patients control mobile devices. However, the large computation power requirement for the ASR system increases implementation costs. To alleviate this problem, this study proposed a convolution neural network (CNN) with a phonetic posteriorgram (PPG) speech feature system to recognize speech commands, called CNN–PPG; meanwhile, the CNN model with Mel-frequency cepstral coefficient (CNN–MFCC model) and ASR-based systems were used for comparison. The experiment results show that the CNN–PPG system provided 93.49% accuracy, better than the CNN–MFCC (65.67%) and ASR-based systems (89.59%). Additionally, the CNN–PPG used a smaller model size comprising only 54% parameter numbers compared with the ASR-based system; hence, the proposed system could reduce implementation costs for users. These findings suggest that the CNN–PPG system could augment a communication device to help dysarthric patients control the mobile device via speech commands in the future.

show abstract

Improving dysarthric speech recognition using empirical mode decomposition and convolutional neural network

Cited by 38 publications

References 15 publications

Stuttered Speech Recognition And Classification Using Enhanced Kamnan Filter And Neural Network

Stuttered Speech Recognition And Classification Using Enhanced Kamnan Filter And Neural Network

Leveraging Classification of Brain Tumour using Deep Learning Architectures

A Speech Command Control-Based Recognition System for Dysarthric Patients Based on Deep Learning Technology

Contact Info

Product

Resources

About