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.

No abstract

The paper reports a solution for the integration of the industrial robot ABB IRB140 with the system for automatic speech recognition (ASR) and the system for computer vision. The robot has the task to manipulate the objects placed randomly on a pad lying on a table, and the computer vision system has to recognize their characteristics (shape, dimension, color, position, and orientation). The ASR system has a task to recognize human speech and use it as a command to the robot, so the robot can manipulate the objects. [Projekat Ministarstva nauke Republike Srbije, br. III44008: Design of Robots as Assistive Technology for the Treatment of Children with Developmental Disorders i br. TR32035: Development of Dialogue Systems for Serbian and other South Slavic Languages

No abstract

No abstract

No abstract

Although the importance of contextual information in speech recognition has been acknowledged for a long time now, it has remained clearly underutilized even in state-of-the-art speech recognition systems. This article introduces a novel, methodologically hybrid approach to the research question of context-dependent speech recognition in human-machine interaction. To the extent that it is hybrid, the approach integrates aspects of both statistical and representational paradigms. We extend the standard statistical pattern-matching approach with a cognitively inspired and analytically tractable model with explanatory power. This methodological extension allows for accounting for contextual information which is otherwise unavailable in speech recognition systems, and using it to improve post-processing of recognition hypotheses. The article introduces an algorithm for evaluation of recognition hypotheses, illustrates it for concrete interaction domains, and discusses its implementation within two prototype conversational agents.

Robotic systems for research and development of factory automation are complex and unavailable for broad deployment in robotic laboratory settings. The usual robotic factory automation setup consists of series of sensors, robotic arms and mobile robots integrated and orchestrated by a central information system. Cloud-based integration has been gaining traction in recent years. In order to build such a system in a laboratory environment, there are several practical challenges that have to be resolved to come to a point when such a system can become operational. In this paper, we present the development of one such system composed of (i) a cloud-based system built on top of open platform for innovation in logistics, (ii) a prototyped mobile robot with a forklift to manipulate pallets in a “factory” floor, and (iii) industrial robot ABB IRB 140 with a customized gripper and various sensors. A mobile robot is designed as an autonomous four Mecanum wheels system with on-board LiDAR and RGB-D sensor for simultaneous localization and mapping. The paper shows a use case of the overall system and highlights the advantages of having a laboratory setting with real robots for the research of factory automation in a laboratory environment. Moreover, the proposed solution could be scaled and replicated in real factory automation applications.