A survey on automatic speech recognition systems for Portuguese language and its variations. (July 2020)

• Record Type: Journal Article
• Authors: Aguiar de Lima, Thales; Da Costa-Abreu, Márjory
• Is Part Of: Computer speech & language. Volume 62(2020)
• Publication Date: 2020-07
• Subjects: Automatic speech recognition -- Text-to-speech -- Speech recognition -- Portuguese -- Survey; Speech processing systems -- Periodicals; Automatic speech recognition -- Periodicals; Computers -- Periodicals; Linguistics -- Periodicals; Speech-Language Pathology -- Periodicals; Traitement automatique de la parole -- Périodiques; Reconnaissance automatique de la parole -- Périodiques; Automatic speech recognition; Speech processing systems; Electronic journals; Periodicals; 006.454
• DOI: 10.1016/j.csl.2019.101055
• Languages: English
• ISSNs: 0885-2308

Abstract

Highlights: Explores the importance of developing techniques for under-resourced languages. Summarises and evaluates studies on Automatic Speech Recognition (ASR) for the Portuguese language. Find literature gaps and draw directions for future research for ASR for the Portuguese language. Abstract: Communication has been an essential part of being human and living in society. There are several different languages and variations of them, so you can speak English in one place and not be able to communicate effectively with someone who speaks English with a different accent. There are several application areas where voice/speech data can be of importance, such as health, security, biometric analysis or education. However, most studies focus on English, Arabic or Asian languages, neglecting other relevant languages, such as Portuguese, which leaves their investigations wide open. Thus, it is crucial to understand the area, where the main focus is: what are the most used techniques for feature extraction and classification, and so on. This paper presents a survey on automatic speech recognition components for Portuguese-based language and its variations, as an understudied language. With a total of 101 papers from 2012 to 2018, the Portuguese-based automatic speech recognition field tendency will be explained, and several possible unexplored methods will be presented and discussed in a collaborative and overall way as our main contribution.