Analysis database Analysis of the sound signal processing

Jurnal Ilmiah Komputer dan Informatika KOMPUTA Edisi. .. Volume. .., Bulan 20.. ISSN : 2089-9033 number5. Picture 6. NPC movements

2.15 Design interface

The design of the interface is needed to represent the true state of the application to be built. Interface design in this application are as follows: Picture 7. Initial appearance Picture 8. Menu chess game Picture 9. Menu how to play

2.16 Testing system

In testing this voice recognition, voice input format used is in accordance with the game of chess. Sound recording format is divided into 3 every voice input by a pause of 1 second. The following sound formats that have been determined as follows: 1. Pawn to be used . 2. Box of letters that will be addressed. 3. Box number to be addressed. There are 10 people consisting of 5 women and 5 men who will be tester with the following criteria: Table 1. Criteria testers Criteria Men age Women age Kid 8 Tahun 8 Tahun Teenage 17 Tahun 18 Tahun Adult 1 34 Tahun 32 Tahun Adult 2 53 Tahun 52 Tahun Oldster 74 Tahun 70 Tahun Based on the above table , it can be seen that it meets the criteria for phase ’s, which includes the stages of human life. Those criteria are divided into two groups, namely men and women. By age’s adjusted criteria. Using the 10 chess pieces that berbebeda movement and represents almost all chess movement. For each test, the player will perform 10x voice input. Table 2. Testing chess piece movements Testing to- Pawn movement Format 1 Format 2 Format 3 1 Pion2 B 4 2 Pion7 G 3 3 Benteng1 A 5 4 Benteng2 H 6 5 Kuda1 D 2 6 Kuda2 G 3 7 Mentri1 B 2 8 Mentri2 E 2 9 Ratu D 6 10 Raja E 2 Here is the result of 10 times of the voice input from 10 different people based on 10 tests with a total of 1,000 data is illustrated by the graph below: Jurnal Ilmiah Komputer dan Informatika KOMPUTA Edisi. .. Volume. .., Bulan 20.. ISSN : 2089-9033 Picture 10. Graph testing accuracy test Based on the above data, it can be calculated the average - average accuracy for each matching method using the formula of the average ’s groups as follows: Table 3. Average’s test of Divide and Conquer Value range test results Mid point �� Frequency �� �� ∗ �� 1 – 10 5.5 11 – 20 15.5 1 15.5 21 – 30 25.5 7 178.5 31 – 40 35.5 2 71 41 – 50 45.5 51 – 60 55.5 61 – 70 65.5 71 – 80 75.5 81 – 90 85.5 91 – 100 95.5 Total 10 265 So can be calculated with the average ’s the data groups and the level of accuracy as follows: � = = . Tabel 4. Average’s test of Euclidean Distance Value range test results Mid point �� Frequency �� �� ∗ �� 1 – 10 5.5 11 – 20 15.5 21 – 30 25.5 2 51 31 – 40 35.5 8 284 41 – 50 45.5 51 – 60 55.5 61 – 70 65.5 71 – 80 75.5 81 – 90 85.5 91 – 100 95.5 Total 10 335 So can be calculated with the average ’s the data groups and the level of accuracy as follows: � = = .

3. Closing

The conclusion is a summary drawn from the discussion made software, where software is created is the application of a chess game using Speech Recognition and compare the application of algorithms Divide and Conquer and Euclidean Distance, while the advice given to the use of these systems is a result of design to be material references to improve performance of this system for the better.

3.1 Conclusion

Based on the analysis that has been carried out on the construction of a chess game application that implements a comparison Divide and Conquer and Euclidean Distance to speech recognition, it can be concluded that the Euclidean Distance is more accurate results to be applied to speech recognition compared to the Divide and Conquer.

3.2 Suggestion

In order for these applications may evolve in the future, the authors suggest some of the following, namely: 1. Improving the results of processing voice signals received by the computer. In order to produce a better value than using Fast Fourier Transform algorithm. 2. Replace all role click the command button into voice, to better reach out to many groups. REFERENCES [1] Park, J.H., Faulkner, J., Schaller, M. 2003. “Evolved Disease-Avoidance Processes And Contemporary Anti-Social Behavior: Prejudical Attitudes And Avoidance Of People With Physical Disabilities. Journal of Nonverbal Behaviour 272”. [2] Fahri Firdausillah, Ika Novita Dewi, Catur Supriyanto. “Sphinx-4 Indonesian Isolated Digit Speech Recognition”. Journal of Theoretical and Applied Information Technology.Volume 53.No.1. Halaman 40. 2013. [3] Nugraha, K. 2011, Aplikasi Perintah Suara Dengan Metode Fast Fourier Transform dan Divide and Conquer pada simulasi rumah pintar. Program Sarjana. Universitas Komputer. 10 20 30 40 50 60 70 80 90 100 Testing accuracy rate Divide and Conquer Euclidean Distance