Speech Technology Progress Based on New Machine Learning Paradigm

Abstract

Speech technologies have been developed for decades as a typical signal processing area, while the last decade has brought a huge progress based on new machine learning paradigms. Owing not only to their intrinsic complexity but also to their relation with cognitive sciences, speech technologies are now viewed as a prime example of interdisciplinary knowledge area. This review article on speech signal analysis and processing, corresponding machine learning algorithms, and applied computational intelligence aims to give an insight into several fields, covering speech production and auditory perception, cognitive aspects of speech communication and language understanding, both speech recognition and text-to-speech synthesis in more details, and consequently the main directions in development of spoken dialogue systems. Additionally, the article discusses the concepts and recent advances in speech signal compression, coding, and transmission, including cognitive speech coding. To conclude, the main intention of this article is to highlight recent achievements and challenges based on new machine learning paradigms that, over the last decade, had an immense impact in the field of speech signal processing.

Effectiveness and Stakeholders’ Perceptions of the Integration of Automated E-Learning Courses into Vocational Education Programmes in Universities in Ukraine

“…Besides, it is consistent with the ideas of both a new paradigm of 21 st -century education (Cunningham, 2019) and robotics-based education (Alimisis, 2020). Interestingly, the recent developments in Speech and Language Technology (SLT), AuthorIT & TutorIT technologies have made it possible to substitute a human-run learning environment losing no instructional quality and ensuring high-cost effectiveness (Delić et al, 2019;Cernak, Asaei & Hyafil, 2018;Scandura, 2010;Scandura, 2016). The letter inspired this study and created a research gap as the study found insufficient limits of information on the use of virtual tutor's assistants in the course delivered on the Moodle platform.…”

N-HANS: A neural network-based toolkit for in-the-wild audio enhancement

“…5 Although N-HANS was implemented to process 16 kHz Waveform Audio File Format (WAV) for input and output, i. e., one of the most standard and broadly used audio formats, input files in other formats or sample rates are also handled through an embedded format conversion based on pysox [4], which internally transfer them into WAV format. 6 Both, single or multiple audio files organised into a directory, can be provided as an input for N-HANS, which achieves an optimal performance with GPU-acceleration 7 but is also capable of running with CPU only. To test N-HANS functionalities, a series of experiments, presented in the following, were conducted.…”

“…Two of the main procedures for enhancing audio and speech are source separation and denoising: The former aims to extract a target audio from a mixture of multiple overlapping signals [66]; the latter attempts to suppress the background noise [49,78]. With the advance of artificial intelligence, neural network based models for audio enhancement have been presented, being particularly efficient in source separation [18,20,35,36,68] and denoising [5,27,32,34,47,60,71] tasks-the performance of classic algorithms is often overtaken by artificial neural networks [7].…”

“…Source code and functionalities are documented at: https://github.com/N-HANS/N-HANS6 Note that for audio files with a quality lower than WAV, such as mp3 in most configurations, the performance of N-HANS may decrease to some degree 7. To carry out any of the considered tasks with a single NVIDIA Titan X Pascal GPU, N-HANS takes about 0.272 seconds to operate one second of audio, i. e., roughly resembling a Real-Time-Factor (RTF) of 0.3.…”

A Review on Artificial Intelligence in Education

“…The continuous improvement of AI technology has been widely used in all walks of life, and the education field is no exception. AI simulates human listening (machine translation, speech recognition) (Delić et al, 2019), speaking (speech synthesis, human-computer dialogue) (Chiba, Nose, Kase, Yamanaka, & Ito, 2019), watching (computer vision, image recognition, text recognition) (Paglen, 2019), thinking (Theorem Proving) (Sarma & Hay, 2017), learning (machine learning, intelligent adaptive learning) (Colchester et al, 2017) and action (robot) (Khandelwal et al, 2017). In particular, AI technologies such as computer vision, natural language processing, and intelligent adaptive learning have changed traditional education and teaching (Yufei, Saleh, Jiahui, & Abdullah, 2020), and have provided universities and teachers with new ideas for teaching reform.…”