Specific SotA

From Control Systems Technology Group
Jump to navigation Jump to search

About Pepper

The following website is the website form the designers from Pepper [1]. Pepper has a speech recognition module and a text to speech system which are available in different languages also dutch and english. With these systems, Pepper is able to hold a conversation but the speech detection is not always reliable. It does not work properly when there is too much noise of if the user talks with an accent. On the website of aldebran, there is a lot information about the languages and programming languages of Pepper [2]. Also, this article specifies the first steps with Pepper (the daily use, settings). It also states how to interact with Pepper [3]. This article [4] gives the use from watson speech in combination with Pepper.

Spoken dialogue processing

This processing has a learning which uses spoken dialogue examples. The dialogue between the user and the robot is processed as the spoken dialogue example. The rule is based on response and reaction of the user. This could be done through inductive learning which uses a genetic algorithm. [5]

Human-Robot dialogue

This article [6] gives a analysis of a human-robot dialogue in the real world. The goal of this analysis is the understanding of the interaction patterns of users. Based on this analysis, implications are described for designing the dialogue of a robot with a user. This article [7] proposes an speech control system for human robot interaction. This control system could understand and translate the intention of human users (human speech commands) into control inputs. This system consists of three parts: a speech recognition system, a control system and a measurement system. This article [8] integrates the perceptual anchoring with a multimodal dialogue in robotics. The goal is to achieve an interaction between robots and humans talking about objects. these objects are located in a system where robots, humans and sensors working together in an environment. They are using the IrisTK dialogue platform which could be runned on a mobile robot device.

Sources

  1. https://developer.softbankrobotics.com/us-en
  2. http://doc.aldebaran.com/2-4/index_dev_guide.html
  3. http://doc.aldebaran.com/2-4/family/pepper_user_guide/index_pepper_user.html
  4. Kumar, H.R.V. (2017). Connecting Pepper Robot with Watson Speech to Text - Java program, retrieved from https://www.ibm.com/developerworks/community/blogs/96960515-2ea1-4391-8170-b0515d08e4da/entry/connecting-pepper-robot-with-watson-speech-to-text-using-java?lang=en
  5. Kimura, Y., Araki, K., Momouchi, Y., & Tochinai, K. (2004). Spoken dialogue processing method using inductive learning with genetic algorithm. Systems and Computers in Japan, 35(12), 67-82. doi:10.1002/scj.10204
  6. Lee, M. K., & Makatchev, M. (2009). How do people talk with a robot? Proceedings of the 27th international conference extended abstracts on Human factors in computing systems - CHI EA 09. doi:10.1145/1520340.1520569
  7. Liu, X., Sam Ge, S., Jiang, R., Goh, C. (2016). Intelligent speech control system for human-robot interaction, Control Conference (CCC), 2016 35th Chinese, doi: 10.1109/ChiCC.2016.7554323
  8. Persson, A., Al Moubayed, S., & Loutfi, A. (2014). Fluent human-Robot dialogues about grounded objects in home environments. Cognitive Computation, 6(4), 914-927. doi:10.1007/s12559-014-9291-y