International Symposium on Telecommunication (IST)

Home / 11th International Symposium on Telecommunication (IST'2024)

Detection and Classification of Stuttering from Text

Authors :

Fateme Moghimi¹ Mehran Yazdi²

1- School of Electrical and Computer Engineering Shiraz University Shiraz 71946, Iran 2- School of Electrical and Computer Engineering Shiraz University Shiraz 71946, Iran

Keywords :

NLP،GAN،adversarial training،BERT،GRU neural network

Abstract :

Stuttering is a language disorder that disrupts the flow of speech, resulting in considerable challenges for those affected by it. Activities such as making phone calls and using new technologies like ASR (Automatic Speech Recognition), can pose a major challenge for someone with stuttering, often making it difficult or even impossible for them to use these tools. Speech-language pathologists and speech therapists manually count the number of stuttering instances in a person's speech to assess the severity of stutter and track its progress over time. However, this method is time-consuming, costly, and prone to errors. Therefore, the automatic evaluation of stuttered speech is crucial for automating the counting and classification of different types of stuttering and is valuable for clinical assessment. In general, this research consists of two parts: detecting stuttered words, and classifying types of stuttering in the DISCO text dataset. The proposed model, CNN-Seq-GAN-BERT, initially identifies stuttered words with 98.6% accuracy. The network's output is subsequently processed by a CNN classifier to distinguish between different stuttering types. The classifier achieved an accuracy of 97.92% on the training data, while the best accuracy on the test data was 80%. Finally, considering that this research is the first to perform text classification on this dataset, the results obtained were compared with the implementation of another method using GRU and transformer neural networks, as well as two machine learning methods, SVM and KNN, to validate the performance of the proposed model.

Papers List

List of archived papers

Examining the legal challenges and loopholes of artificial intelligence by considering the upstream documents of the country

Mohammad Hadi Bokaei - Elham Rafati - Davood Heidari Kani - Farbod Rabizadeh Fard - Ali Derakhshani - Iman Derakhshani - Nooshin Bodaghi - Reza Baderestani - Neda Nedaei

Design and Analysis of a THz Topological Frequency-Tunable Waveguide based on Valleytronics

Mohammadreza Moeini - Vahid Ahmadi

A Low-Cost Passive Thermal IR Imaging System for Automated Hidden Object Detection Using AI

Mohammad Javad Amiri - Saman Ghazanfari - Mohammad Parsa Mirzaei Sarmad - Mohammad Fakharzadeh

Cryptocurrency volatility prediction based on price, return and volatility cross-correlation using LSTM

Masoud Omidvari Abarghouie - Sasan H. Alizadeh - Ahmad Khademzadeh

Classification of Congestive Heart Failure Disease, Arrhythmias, And Normal Heart Rhythms Using Deep Learning

Nafiseh Batavani - Mohammad Reza Yousefi

Analytical Performance model for Hyperledger Fabric for Communication Technology use cases

Amin Ghalami Osgouei - Touraj Shabanian - Mehdi Samie - Katayoun Bamasian

Performance Comparison of Internet Domain Ranking Techniques

Davood Maleki - Pejman Goudarzi - Alireza Mansouri

Compact Wideband Circularly Polarized Printed Monopole Antenna

Mubarak Sani Ellis - Javad Nourinia - Changiz Ghobadi - Keyhan Hosseini - Farzad Alizadeh - Bahman Mohammadi

Analyzing the Challenges of Applying Artificial Intelligence in the Field of Employment and Offering Solutions to Improve Skills and Workforce Training

Azam sadat Mortazavi Kahangi - Hassan Yeganeh - Anita Hadizadeh

The cost of renting a fiber network from one operator to another in Iran

Reza Mehri - Davoud Ranjbarrafie - Elham Ziaeipour