Comprehensive Review Paper on SpeechEase: AI-Based Multilingual Speech Therapy System

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Speech and pronunciation difficulties can affect communication, learning, academic participation, and social interaction.Conventional speech assessment and therapy generally depend on trained professionals and repeated practice sessions, which may not always be easily accessible to all users.Recent developments in Artificial Intelligence (AI), Automatic Speech Recognition (ASR), machine learning, deep learning, and speech signal processing have created opportunities for automated speech assessment and personalized pronunciation practice.This paper reviews AI-assisted speech technologies with emphasis on multilingual speech processing, speech-sound analysis, personalized feedback, and progress monitoring.It follows a focused semi-systematic/narrative review approach and discusses selected literature on automatic speech analysis, personalized pronunciation correction, and articulatory analysis.The reviewed work indicates that AI may support speech practice outside continuous clinician supervision; however, language coverage, phonemespecific evaluation, dataset availability, and integration of assessment with longitudinal practice remain challenges.The review identifies a research gap in combining multilingual pronunciation assessment, weak-sound or word identification, targeted practice, repeated re-assessment, and day-wise progress tracking in one platform.An initial proposed framework is outlined as a future development direction.No application has been implemented or experimentally evaluated at this stage.The proposed system is intended to support practice and early intervention, not to replace professional speech-language assessment or clinical therapy.