**6. Tutoring strategies**

Although our MITS can perfectly detect the attention information and affective state, it is also important to let the agent tutor know what to do with the information. As good human tutors can effectively adapt to the attention information and affective state of students, the most obvious way to learn about how to adapt the attention information and affective state of student is to learn from the human tutors. Sarrafzadeh videoed several tutors as they tutored students individually and a coding scheme was developed to extract data from each tutoring video to describe the behaviors, facial expressions and expression intensities of students and tutors. Tutoring actions are guided by a case-based method for adapting to student states that recommends a weighted set of tutor actions and expressions (Sarrafzadeh

Speech is the easiest way to convey intention, and it is one of the fundamental methods of conveying emotion, on a par with facial expression. In this paper, the variety rule of prosodic features containing pitch frequency (F0), energy and velocity are concluded by analyzing emotional speech in our Emotional Speech Database. The autocorrelation function (ACF) method based on Linear Predictive Coding (LPC) and wavelet transform approach are employed to extract the F0 and tone respectively. Then prosodic features regulation is set up by utilization of Pitch Synchronous OverLap Add (PSOLA) and the original peace speeches are transformed into appointed emotional speech, including happy, anger, surprise and sad,

based on the rules and regulation. Figure 8 illustrates the work flow of our approach.

 **Pre-process.** Include noise elimination, pre-emphasis and amplitude normalization **LPC analyze.** Partition the original speech into frame, take LPC analysis of each frame,

**F0 extraction.** Get the F0 through the autocorrelation analysis of the residual function

**Surd and sonant separation.** Do the surd and sonant separation according to first order

**PSOLA**. Using PSOLA technology to transform original speech into appointed

**Post-process.** De-emphasis, i.e. do the anti-operation of pre-emphasis in pre-process to

Although our MITS can perfectly detect the attention information and affective state, it is also important to let the agent tutor know what to do with the information. As good human tutors can effectively adapt to the attention information and affective state of students, the most obvious way to learn about how to adapt the attention information and affective state of student is to learn from the human tutors. Sarrafzadeh videoed several tutors as they tutored students individually and a coding scheme was developed to extract data from each tutoring video to describe the behaviors, facial expressions and expression intensities of students and tutors. Tutoring actions are guided by a case-based method for adapting to student states that recommends a weighted set of tutor actions and expressions (Sarrafzadeh

 **Glottal Closure Instances (GCI).** determine the GCI according to the F0 extraction **Tone modification.** Extract the tone information using wavelet transform, modify the tone information according to the target-emotion; adopting inverse wavelet transform

get the LPC residual function and first order reflection coefficient

reflection coefficient, signal energy and frequency extraction result

**5.2 Emotional speech synthesis** 

Fig. 8. Work flow of the emotional speech synthesis.

and the F0 profile curve of the original speech

to get the F0 curve

emotional speech

**6. Tutoring strategies** 

restore speech effect

& Alexander, 2008). However, this approach has two main shortcomings: firstly, the video only recorded three human tutors' behaviors and the students' reactions to these behaviors are not considered. What we want to get is the behaviors that can motivate the students, rather than arouse the students' averseness; and secondly, the coding scheme can not apply to the attention information and speech, text communication. In this paper, we use the traditional questionnaire to get the "optimal reaction" of the tutor towards the learner's attention information and affective state. The critical observation is that every excellent teacher has commonsense of the kind we want to give our agent tutor. If we can find good ways to extract commonsense from human tutor by prompting them, asking them questions, presenting them with lines of reasoning to confirm or repair, and so on, we may be able to accumulate many of the knowledge structures needed to give our agent tutor the capacity for commonsense reasoning for student's attention information and affective state. So we built a system called Human Tutor Commonsense make it easy for human tutors to collaborate to construct a database of commonsense knowledge. We invited more than 100 excellent teachers to log on our system to build the database. Then, a group of 50 students were asked to evaluate how much they satisfied with these commonsense, on a scale from 1 (strongly dissatisfied) to 5 (strongly satisfied). Then we chose two commonsense with highest mean score as the "optimal reaction" for each situation these questions described. Based on the commonsense we obtained, MITS can be represented as a dynamic network as

shown in Figure 9. Whenever the student's pedagogical state or attention information or affective state is changed, the following events are happen:


Fig. 9. Dynamic network for MITS.

## **7. Conclusion**

This paper debuts a Multimodal Intelligent Tutoring Systems. Attention information detection and affective state detection are carried out. Meanwhile, the system adapts to the

**7** 

*Portugal* 

**Using the Smith Chart in an** 

*Universidade de Aveiro, Instituto de Telecomunicações* 

*Instituto Superior de Engenharia de Lisboa, Instituto de Telecomunicações* 

Prior to the advent of digital computers and calculators, engineers developed all sorts of aids (tables, charts, graphs) to facilitate their calculations for design and analysis in different areas in particular for line transmission problems. To reduce the tedious manipulations involved in calculating the characteristics of transmission lines, graphical tools have been developed. The Smith chart is the most commonly used of these graphical techniques. It is basically a graphical indication of the impedance change along a transmission line as one moves along it. It becomes easy to use after a small amount of experience. We will first explain how the Smith chart is constructed and then how to use it to calculate transmission line characteristics such as: the reflection coefficient (), the Voltage Standing Wave Ratio (VSWR), the impedance along the line (Z(d)), the maximum and minimum voltage localization and impedance matching. For the majority of these Smith chart applications

Since the main topic of this book is concerned with e-learning, the aim of this chapter is to help the reader understand and learn how to use the Smith chart, following step by step procedure based on *MATLAB* scripts that will be available for download and should be used when reading this chapter. This approach should teach the students how to solve several kinds of transmission line problems by themselves, in a paper chart using a pencil, a

*MATLAB* scripts are a good tool to help students better understand the Smith chart and explain, step by step, several procedures related to transmission line problems, (Mak &

The goals of the chapter are to explain the reasons why using and understanding the Smith chart is still important nowadays, despite the present generalization of personal computers and powerful calculators. It is easy to plug a few numbers into a program and have it spit out solutions. When the solutions are complex and multifaceted, having a computer to do the grunt work is especially handy. However, knowing the underlying theory and principles that have been ported to computer platforms, and where they came from, makes the engineer or designer a more well-rounded and confident professional, and makes the results more reliable. Moreover it is interesting to note that these kinds of graphical tools are still useful nowadays. For example some types of modern laboratory equipment, such as

lossless lines will be assumed, although this is not absolutely required.

**1. Introduction**

ruler and a compass.

Sundaram, 2008), (Pereira & Pinho, 2010).

**E-Learning Approach** 

José R. Pereira and Pedro Pinho

student via an emotionally expressive agent tutor "Alice" through facial expression and synthetic emotional speech. Tutoring actions are guided by a case-based method that recommends a set of tutor actions and expressions for adapting to student states. The data that this case-based program uses were generated from questionnaires presented to human teachers.

In future work, it is necessary that the accuracy of emotion recognition and classification algorithm should be improved. Meanwhile, the MITS will be extended to integrate information from other sources including posture recognition and physiological channels such as pressure. We hope to evaluate the effectiveness of our system in a range of learning situations including more both young and adult learners. The test will provide more important directions for improvements to be made in the next version of MITS.
