HIDDEN MARKOV MODEL SEDERHANA UNTUK APLIKASI PENGENALAN ISYARAT PENUTUR

Authors

  • Septiantar Tebe Nursaputro
  • Dwi Dharma Arta Kusuma
  • Risanuri Hidayat
  • Oyas Wahyunggoro

DOI:

https://doi.org/10.32497/orbith.v19i3.5265

Keywords:

passwords, Mean Square Error, Hidden Markov Model

Abstract

Umumnya, orang memasukkan sandi dengan mengetik data mereka. Hal ini berisiko karena dapat terjadi kemungkinan untuk memberikan data yang sama kepada orang lain dan tidak ada ciri unik data tersebut. Salah satu metode untuk memecahkan masalah menggunakan pengenalan dengan klasifikasi Markov Model tersembunyi. Ada beberapa teknik untuk mendapatkan ciri pada pengenalan tutur; salah satunya adalah mengambil data langsung. Pertama, ekstraksi ciri dilakukan menggunakan transformasi wavelet. Proses belanjut ke Model Markov tersembunyi dimana nilai probabilitas ditentukan berdasarkan toleransi kesalahan dari suatu data percobaan. Nilai kesalahan didapat dengan menggunakan Mean Square kesalahan (MSE). Metode ini dapat mendeteksi ucapan dengan intonasi berbeda dengan akurasi 100% dalam jarak 25 sampai 40 cm, tetapi tidak untuk dialek berbeda.
Kata Kunci : kata sandi, Mean Square Error, Hidden Markov Model

Generally, people enter the password by typing their data. This is risky because of the probability to give similar data to others and there is no unique feature of data. One ofthe methods to solve the problem is using speech recognition with classification of Hidden Markov Model. There are some techniquesto get the features in speech recognition; one of them is direct data mining. Firstly, feature extraction was done using wavelet transform. The process was continued to Hidden Markov Model which the input probabilitywas determined based on the error tolerance of experiment data. The error value was got using Mean Square Error (MSE). This proposed method can detect the speech with different intonation with the accuracy of 100% within the distance of 25 to 40 cm, but not for different dialect.
Keywords: passwords, Mean Square Error, Hidden Markov Model

References

Apandi, T. H. (2016). Prediksi Trafik Video Dengan Menggunakan. 101”“104.

Are, G. P. B., Sitorus, S. H., Prof, J., Hadari, H., & Pontianak, N. (2020). Prediksi Nilai Tukar Mata Uang Rupiah Terhadap Dolar Amerika Menggunakan Metode Hidden Markov Model. Coding : Jurnal Komputer Dan Aplikasi, 08(01), 44”“54.

Dwi Ramadhan, H. (2017). Analisis Pemeliharaan Prediktif Transformator Daya di PT. PLN GI Blimbing Malang Dengan Metode Markov. 9”“10.

Kuswoyo, R., Dur, S., & Cipta, H. (2023). Penerapan Proses Stokastik Markov Chain Dalam Pengendalian Persediaan Produksi Kelapa Sawit di Perkebunan Nusantara IV Sumatera Utara. G-Tech: Jurnal Teknologi Terapan, 7(2), 429”“438. https://doi.org/10.33379/gtech.v7i2.2025

Marganingsih, M., & Rosidin, O. (2023). Dengan Kelainan Dengar Konduktif. 19, 28”“39.

Masriastri, I. G. A. K. Y. (2018). Perpustakaan dan masyarakat informasi. Al-Maktabah, 3(2), 72”“83.

Meiyanti, R., & Mestika Sandy, C. L. (2021). Pendeteksi Pengenalan Emosi Pada Manusia Menggunakan Hidden Markov Model Dan Bidirectional Associative Memory Dengan Suara. Jurnal Tika, 6(03), 231”“237. https://doi.org/10.51179/tika.v6i03.75 6

Prastowo, B. N., Putro, N. A. S., Dhewa, O.

A., & Yusuf, A. M. H. (2019). Pengenalan Personal Menggunakan Citra Tampak Atas pada Lingkungan Cashierless Strore. Jurnal Buana Informatika, 10(1), 19. https://doi.org/10.24002/jbi.v10i1.1779

Suarnatha, I. P. D., Agus, I. M., & Gunawan, O. (2022). Jurnal Computer Science and Information Technology ( CoSciTech ) manusia. CoSciTech, 3(2), 73”“80.

Wicaksono, A. T., & Fatimah, T. (2018). Sistem Penilaian Online Menggunakan Keamanan One Time Password Dengan Algoritma Sha 512 Berbasis Web. Skanika, 1(3), 938”“943.

Yahya, N. I., & Amini, S. (2018). Pengimplementasian One Time Password Dan Notifikasi Email Menggunakan Fungsi Hash SHA- 512 Berbasis Web Pada SMK Cyber Media. Skanika, 1(2), 745”“750. https://jom.fti.budiluhur.ac.id/index.php/SKANIKA/article/view/285

Downloads

Published

2023-12-21

Issue

Vol. 19 No. 3 (2023): November 2023

Section

Articles

License

Authors who publish with this journal agree to the following terms:

Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.

Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.

Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).