Speech Stuttering Detection and Removal Using Deep Neural Networks

Shaswat Rajput, Ruban Nersisson, Alex Noel Joseph Raj, A. Mary Mekala, Olga Frolova, Elena Lyakso

Результат исследований: Публикации в книгах, отчётах, сборниках, трудах конференцийстатья в сборнике материалов конференциинаучнаярецензирование


There are more than 70 million people worldwide who suffer from stuttering problems. This will affect the confidence of public speaking in people who suffer from this issue. To solve this problem many people take therapy sessions but the therapy sessions are a temporary solution, as soon as they leave therapy sessions this problem might arise again. This work aims to use state of the art machine learning algorithms that have improved over the past few years to solve this problem. We have used the dataset from UCLASS archives which provide the data for stuttered speech in.wav format with time-aligned transcriptions. We have tried different algorithms and optimized our model by hyper parameter tuning to maximize the model’s accuracy. The algorithm is tested on random speech data with low to heavy stuttering from the same dataset, and it is observed that there is significant reduction in the Word Error Rate (WER) for most of the test cases.

Язык оригиналаанглийский
Название основной публикацииProceedings of the 11th International Conference on Computer Engineering and Networks
РедакторыQi Liu, Xiaodong Liu, Bo Chen, Yiming Zhang, Jiansheng Peng
ИздательSpringer Nature
Число страниц9
ISBN (печатное издание)9789811665530
СостояниеОпубликовано - 2022
Событие11th International Conference on Computer Engineering and Networks, CENet2021 - Hechi, Китай
Продолжительность: 21 окт 202125 окт 2021

Серия публикаций

НазваниеLecture Notes in Electrical Engineering
Том808 LNEE
ISSN (печатное издание)1876-1100
ISSN (электронное издание)1876-1119


конференция11th International Conference on Computer Engineering and Networks, CENet2021

Предметные области Scopus

  • Промышленная технология и станкостроение


Подробные сведения о темах исследования «Speech Stuttering Detection and Removal Using Deep Neural Networks». Вместе они формируют уникальный семантический отпечаток (fingerprint).