| １． |
Outline |
|
We will study voice recognition as an example of the framework of pattern recognition processing systems. Next, we learn the theorem of pattern recognition focusing on character recognition which regards to the following concerns: Bayes’ theorem and estimation of probability density function based on statistical pattern recognition theory, discrimination functions which greatly affect recognition result, some non-linear recognizers such as neural networks, and recent research results of learning and generalization.
This course is related to DP1.
|
| ２． |
Objectives |
|
The aim of the course is to comprehend the automatic pattern recognition by computer from the viewpoint of expanding information processing capability of humans—from signal processing to encoding, categorization, recognition and understanding—using examples such as application of voice recognition systems.
Next, students will study the theorem of pattern recognition focusing on character recognition which regards to the following concerns: Bayes’ theorem and estimation of probability density function based on statistical pattern recognition theory, discrimination functions which greatly affect recognition result, some non-linear recognizers such as neural networks, and recent research results of learning and generalization.
|
| ３． |
Grading Policy |
|
Assessed by the reports.
The learners can get feedback from the reports commented by the professor.
|
| ４． |
Textbook and Reference |
|
The learners are provided with some learning materials as necessary.
|
| ５． |
Requirements (Assignments) |
|
This course is middle level among pattern recognition related subjects in graduate school. The participants had better learn the image processing and the voice processing at the next stage of this course. The learners should review information algebra, coding theory, probability theory, information theory, and mathematical programming.
The learners must read the materials carefully then confirm concept of the keyword of each class before class (1.5 hours), and read some related papers after the class (1.5 hours).
|
| ６． |
Note |
|
|
| ７． |
Schedule |
|
１. Human vs. computers: the difference and similarity among two
|
２. Sensing and its characteristics: perception from the outside, transmission, and kinds and characteristics of sensor
|
３. Sensing and pattern recognition: difference, similarity and principle of pattern recognition between computers and human
|
４. Receptor and physical characteristics of sound signal: reception and processing for the sense of hearing sound signal
|
５. Information processing of auditory sense: structure, behavior and signal processing of the outer, the middle and the inner ear
|
６. Speech recognition: difference, similarity and principle of voice recognition between computers and human
|
７. Memory and learning for pattern recognition
|
８. Voice recognition technologies: voice recognition, voice synthesis, voice quality conversion and other voice processing technologies
|
９. Statistical pattern recognition theory: Bayes' theorem, estimation for probability density function
|
10. Discrimination function (1): multilayered neural network and other non-linear discrimination functions
|
11. Discrimination function (2): multilayered neural network and other non-linear discrimination functions
|
12. Learning and generalization (1): recent research result and theme
|
13. Learning and generalization (2): recent research result and theme
|
14. Technologies for hand-written character recognition (1)
|
15. Technologies for hand-written character recognition (2)
|
|
| １． |
Outline |
|
We will study voice recognition as an example of the framework of pattern recognition processing systems. Next, we learn the theorem of pattern recognition focusing on character recognition which regards to the following concerns: Bayes’ theorem and estimation of probability density function based on statistical pattern recognition theory, discrimination functions which greatly affect recognition result, some non-linear recognizers such as neural networks, and recent research results of learning and generalization.
This course is related to DP1.
|
| ２． |
Objectives |
|
The aim of the course is to comprehend the automatic pattern recognition by computer from the viewpoint of expanding information processing capability of humans—from signal processing to encoding, categorization, recognition and understanding—using examples such as application of voice recognition systems.
Next, students will study the theorem of pattern recognition focusing on character recognition which regards to the following concerns: Bayes’ theorem and estimation of probability density function based on statistical pattern recognition theory, discrimination functions which greatly affect recognition result, some non-linear recognizers such as neural networks, and recent research results of learning and generalization.
|
| ３． |
Grading Policy |
|
Assessed by the reports.
The learners can get feedback from the reports commented by the professor.
|
| ４． |
Textbook and Reference |
|
The learners are provided with some learning materials as necessary.
|
| ５． |
Requirements (Assignments) |
|
This course is middle level among pattern recognition related subjects in graduate school. The participants had better learn the image processing and the voice processing at the next stage of this course. The learners should review information algebra, coding theory, probability theory, information theory, and mathematical programming.
The learners must read the materials carefully then confirm concept of the keyword of each class before class (1.5 hours), and read some related papers after the class (1.5 hours).
|
| ６． |
Note |
|
|
| ７． |
Schedule |
|
１. Human vs. computers: the difference and similarity among two
|
２. Sensing and its characteristics: perception from the outside, transmission, and kinds and characteristics of sensor
|
３. Sensing and pattern recognition: difference, similarity and principle of pattern recognition between computers and human
|
４. Receptor and physical characteristics of sound signal: reception and processing for the sense of hearing sound signal
|
５. Information processing of auditory sense: structure, behavior and signal processing of the outer, the middle and the inner ear
|
６. Speech recognition: difference, similarity and principle of voice recognition between computers and human
|
７. Memory and learning for pattern recognition
|
８. Voice recognition technologies: voice recognition, voice synthesis, voice quality conversion and other voice processing technologies
|
９. Statistical pattern recognition theory: Bayes' theorem, estimation for probability density function
|
10. Discrimination function (1): multilayered neural network and other non-linear discrimination functions
|
11. Discrimination function (2): multilayered neural network and other non-linear discrimination functions
|
12. Learning and generalization (1): recent research result and theme
|
13. Learning and generalization (2): recent research result and theme
|
14. Technologies for hand-written character recognition (1)
|
15. Technologies for hand-written character recognition (2)
|
|
|