Experiments With Fast Fourier Transform, Linear Predictive and Cepstral Coefficients in Dysarthric Speech Recognition Algorithms Using Hidden Markov Model

2005 ◽  
Vol 13 (4) ◽  
pp. 558-561 ◽  
Author(s):  
P.D. Polur ◽  
G.E. Miller
Keyword(s):  
Fourier Transform ◽  
Speech Recognition ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Fast Fourier Transform ◽  
Hidden Markov ◽  
Recognition Algorithms ◽  
Dysarthric Speech ◽  
Cepstral Coefficients

scholarly journals Speech Recognition dengan Hidden Markov Model untuk Pengenalan dan Pelafalan Huruf Hijaiyah

2019 ◽  
Vol 5 (1) ◽  
pp. 19
Author(s):  
Qothrun Nada ◽  
Cahya Ridhuandi ◽  
Puji Santoso ◽  
Dwi Apriyanto
Keyword(s):  
Fourier Transform ◽  
Speech Recognition ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Fast Fourier Transform ◽  
Euclidean Distance ◽  
Markov Models ◽  
Hidden Markov ◽  
Learning To Read ◽  
Machine Interface

<p><em>Abstrak</em><strong> </strong>- <strong>Pelajaran utama dalam membaca Al Qur'an adalah mengenali dan </strong><strong>melafalkan</strong><strong> huruf-huruf Hijaiyah. Beberapa fakta menunjukkan bahwa pengucapan yang salah dapat memengaruhi makna secara har</strong><strong>a</strong><strong>fiah. <em>Speech Recognition</em>, sebagai teknologi saat ini, dapat digunakan untuk memeriksa kesalahan dalam melafalkan surat Hijaiyah melalui pengenalan suara atau ucapan. Itu dapat dikonversi menjadi data yang dapat dipahami oleh sistem. Tujuan dari penelitian ini adalah untuk menerapkan <em>Speech Recognition</em> dengan <em>Hidden Markov Model</em> untuk pelafalan huruf Hijaiyah ketika belajar membaca Alquran. Pengenalan ucapan dan Model Hidden Markov dilakukan untuk mengembangkan sistem antar</strong><strong> </strong><strong>muka mesin berbasis suara. Dalam penelitian ini juga menggunakan metode <em>Fast Fourier Transform</em> (FFT) untuk mengekstraksi sifat. <em>Hidden Markov Model</em> (HMM) yang digunakan dalam proses pelatihan. Juga, menghasilkan karakteristik khusus untuk setiap huruf Hijaiyah. Dan kemudian, <em>Euclidean Distance</em> (ED) untuk klasifikasi akhir dalam mendeteksi pelafalan huruf Hijaiyah. Hasil penelitian menunjukkan bahwa hasil tes huruf Hijaiyah pada tingkat akurasi yang sama adalah 100%, sedangkan pengujian huruf yang berbeda adalah 54,6%. Dengan demikian, penelitian ini akan memberikan kontribusi kepada siswa yang sedang belajar membaca Al-Qur'an untuk dapat mengenali dan me</strong><strong>lafalkan</strong><strong> huruf-huruf Hijaiyah</strong><strong><em>.</em></strong></p><p><em>Abstract</em> – <strong>The main lesson in reading the Al Qur'an is recognizing and reciting the letters Hijaiyah. Some facts show that incorrect pronunciation can affect meaning literally. Speech Recognition, as the current technology, can be used to check the mistakes in pronouncing the Hijaiyah's letter through recognizing the voice or speech. It can convert into data that can be understood by the system. The purpose of this study is to implement Speech Recognition with Hidden Markov Model for Hijaiyah letter pronunciation when learning to read the Qur'an. Speech recognition and Hidden Markov Models were carried out to develop a sound-based machine interface system. In this study also used the Fast Fourier Transform (FFT) method to extract traits. Hidden Markov Model (HMM) used in the training process. Also, produced the especially characteristics for each letter of Hijaiyah. And then, Euclidean Distance (ED) for the final classification in detecting Hijaiyah letter pronunciation. The results of the study show that the results of the Hijaiyah letter test on the same level of accuracy are 100%, while the testing of different letters is 54.6%. Thus, this study will contribute to students who are learning to read Al-Qur'an to be able to recognize and recite the Hijaiyah letters</strong><strong><em>.</em></strong></p><p><strong><em>Keywords</em></strong> -  <em>Speech Recognition, Hidden Markov Model, Recognizing, Reciting, Letter Hijaiyah </em></p>

Implementation of Android Based Speech Recognition for Indonesian Geography Dictionary

2016 ◽  
Vol 7 (2) ◽  
pp. 76-82
Author(s):  
Hugeng Hugeng ◽  
Edbert Hansel
Keyword(s):  
Speech Recognition ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Hidden Markov ◽  
Spoken Word ◽  
Silent Condition ◽  
Word Detection ◽  
Index Terms ◽  
To Receive ◽  
Noisy Condition

We have built an application of speech recognition for Indonesian geography dictionary based on Android operating system, named GAIA. This application uses a smartphone as a device to receive input in the form of a spoken word from a user. The approach used in recognition is Hidden Markov Model which is contained in the Pocketsphinx library. The phonemes used are Indonesian phonemes’ rule. The advantage of this application is that it can be used without internet access. In the application testing, word detection is done with four conditions to determine the level of accuracy. The four conditions are near silent, near noisy, far silent, and far noisy. From the testing and analysis conducted, it can be concluded that GAIA application can be built as a speech recognition application on Android for Indonesian geography dictionary; with the results in the near silent condition accuracy of word recognition reaches an average of 52.87%, in the near noisy reaches an average of 14.5%, in the far silent condition reaches an average of 23.2%, and in the far noisy condition reaches an average of 2.8%. Index Terms—speech recognition, Indonesian geography dictionary, Hidden Markov Model, Pocketsphinx, Android.

Sign in / Sign up

Export Citation Format

Share Document