Showing posts with label Machine Sound Separation. Show all posts
Showing posts with label Machine Sound Separation. Show all posts

Saturday, May 23, 2015

Interaural Time Difference (ITD)

ITD merupakan kepanjangan dari interaural time difference,  yakni perbedaan waktu tempuh sumber suara ke telinga kiri dan telinga kanan. Apa pentingnya ITD ini? Penting sekali, dengan ITD kita bisa memperkirakan dan mengetahui sumber suara tanpa melihatnya. Coba tutup mata anda, dan dengarkan suara yang ada, kemudian perkirakan di mana letak sumber suara tersebut. Sekarang buka mata anda, dan crosscheck posisi sumber suara yang anda perkirakan tadi, persis dengan perkiraan anda.

Ilustrasi 1: Sumber suara pada 0 derajat dan 30 derajat. Pada 0 derajat ITD adalah 0, sedang pada 30 derajat ITD dapat dicari pada gambar/grafik 3 (sumber: Thesis BT Atmaja [1], dok pribadi)

Teori Duplex
Teori tentang ITD ini pertama kali dikemukakan oleh Lord Rayleigh (1907) yang menyatakan bahwa perbedaan suara yang sampai pada telinga dapat dicari/dihitung,
provided evidence that timing differences between the ears were detectable
Teori duplex Lord Rayleigh ini mengandung dua dasar pengukuran sensitivitas pada ILD (Interaural level difference, "beda level") dan sensitivitas pada ITD ("beda waktu") sebagai berikut.

Monday, October 31, 2011

Machinery fault diagnosis using independent component analysis (ICA) and Instantaneous Frequency (IF)

Machine condition monitoring plays an important role in industry to ensure the continuity of the process. This work presents a simple and yet, fast approach to detect simultaneous machinery faults using sound mixture emitted by machines. We developed a microphone array as the sensor. By exploiting the independency of each individual signal, we estimated the mixture of the signals and compared time-domain independent component analysis (TDICA), frequency-domain independent component analysis (FDICA) and Multi-stage ICA. In this research, four fault conditions commonly occurred in industry were evaluated, namely normal (as baseline), unbalance, misalignment and bearing fault. The results showed that the best separation process by SNR criterion was time-domain ICA. At the final stage, the separated signal was analyzed using Instantaneous Frequency technique to determine the exact location of the frequency at the specific time better than spectrogram. 

Full paper is available in IEEExplore