Ses Sinyalinde Gürültü Saptama İçin Özgün Bir Yaklaşım

Yıl 2018, Cilt: 1 Sayı: 1, 31 - 38, 18.12.2018

Ramazan Çolak Rafet Akdeniz

Öz

Ses etkinliği algılama, genelde konuşma tanıma, konuşma sıkıştırma gibi konuşma işleme süreçlerinin başında kullanılan bir uygulamadır. Konuşma sesinin olup olmadığını tespit etmede kullanılır ve buna göre uygulamanın devamına yön verir. Sesin varlığını tespit etmede kullanılan belli başlı özellikler vardır. Kullanılan özelliklerin fazla olması algoritmanın verimliliği ile doğrudan ilişkilidir. Klasik VAD algoritmaları genelde STE kullanılarak oluşturulduğundan, düşük sinyal gürültü oranı değerlerinde çok hassastır, bu yüzden istenilen sonuçları veremeyebilir. Çözümde kullanılan özellikler için gerçek zamanlı sesler kullanarak sesli bölge ve gürültülü bölge ayırt edilmeye çalışılmıştır. Bu çalışmada sesin varlığını tespit etmek için STE, periyodiklik ve Spektral düzlük gibi üç özellik kullanılmıştır, kullanılan bu üç özellik ile düşük SNR değerlerinde de istenilen sonuçlar elde edilmiştir. Bu yöntemin, özellikle düşük SNR değerlerinde klasik metotlara göre daha iyi performans elde ettiği gözlemlenmiştir.

Anahtar Kelimeler

En baskın frekans bileşeni, Kısa zamanlı enerji, Spektral düzlük, Ses etkinliği algılama

A Novel Approach to Noise Reduction in Audio Signal

Öz

Voice activity detection is often used at the beginning of speech processes, such as speech recognition, and speech compression. It is used to detect the presence of a speaking voice, and it directs the execution of the application accordingly. There are certain features that are used to detect the presence of the voice. The efficiency of the algorithm is directly related to the number of the features used. Since the classical Voice activity detection algorithms are usually developed using Short Time Energy, they are very sensitive to the low signal-to-noise ratio values, therefore they may not provide the desired results. The parts with a speech and the parts with noise were attempted to be distinguished by using real-time sounds for the features used in the solution. In this study, three features, such as Short Time Energy, Periodicity, and the Spectral Flatness, were used to detect the voice. The desired results have been obtained by using these three features, even at low SNR values. This method has been observed to achieve better performance especially at low SNR values than conventional methods.

Anahtar Kelimeler

Short-time energy, Spectral flatness, Most dominant frequency component, Voice activity detection

Kaynakça

• [1] M. H. Moattar and M. M. Homayounpour, “A Simple But Efficient Real-Time Voice Activity Detection Algorithm”, 17th EUSIPCO, pp. 2549-2553, 2009.
• [2] K. Sakhnov, E. Verteletskaya, B. Simak, “Low Complexity Voice Activity Detector Using Periodicity And Energy Ratio”, 16th International Conference on Systems, Signals and Image Processing IEEE, pp. 1-5, 2009.
• [3] E. Verteletskaya, K. Sakhnov, “Voice Activity Detection for Speech Enchancement Applications”, ACTA POLYTECHNICA, Vol.50, No.4, 2010.
• [4] M. H. Moattar, M. M. Homayounpour, N.K. Kalantari “A New Approach For Robust Realtime Voice Activity Detection Using Spectral Pattern”, International Conference on Acoustic, Speech and Signal Processing IEEE, pp. 4478-4481, 2010.
• [5] Y. K Bharath, S. Veena, K. V. Nagalakshmi, Manjunath Darshan, Rohini Nagapadma, “Development of Robust VAD Schemes for Voice Operated Switch Application in Aircrafts”, 2nd International Conference on Applied and Theoretical Computing and Communication Technology (iCATccT), pp. 191-195, 2016.
• [6] T. H. Zaw, N. War “The Combination of Spectral Entropy, Zero Crossing Rate, Short Time Energy, and Linear Prediction Error for Voice Activity Detection”, International Conference of Computer and Information Technology (ICCIT), pp. 1-5, 2017.
• [7] N. Lezzoum, G. Gagnon, J. Voix, “Voice Activity Detection System for Smart Earphones”, IEEE Transaction on Consumer Electronics, Vol. 60, pp. 737-744, 2014.
• [8] M. Kumari, I, Ali, “An Efficient Un-Supervised Voice Activity Detector for Clean Speech”, International Conference on Communication, Control and Intelligent Systems (CCIS), pp. 227-232, 2015.
• [9] A. Pasad, K. Sabu, P. Rao, “Voice Activity Detection for Children’s Read Speech Recognition in Noisy Conditions”, Twentythird National Conference on Communications (NCC), pp. 1-6, 2017.
• [10] J. Pang, “Spectrum Energy Based Voice Activity Detection”, IEEE 7th Annual Computing and Communication Workshop and Conference (CCWC), pp. 1-5, 2017.
• [11] I. Almajai, B. Milner, “Using Audio-Visual Features for Robust Voice Activity Detection in Clean and Noise Speech”, 16th European Signal Processing Conference (EUSIPCO), pp. 1-5, 2008.
• [12] P. Pollak, P. Sovka and J. Uhlir, “Noise Suppression System For A Car”, Third European Conference on SpeechCommunication and Technology, 3rd European Conference on Speech Communication and Technology-EUROSPEECH pp. 1073-1076, 1993.
• [13] K. Sakhnov, E. Verteletskaya, B. Simak, “Dynamical Energy-Based Speech/Silence Detector for Speech Enhancement Applications”, in Proc. of the World Congress on Engineering, vol. 1, pp. 801, 2009.
• [14] IEEE Recommended Practice for Speech Quality Measurements. IEEE Trans. Audio and Electroacoustics, Vol. 17, pp. 225-246, 1969.
• [15] http://ecs.utdallas.edu/loizou/speech/noizeus/