Interactive controller for audio object localization and automatic thumbnail music generator
In this demonstration, we provide a new interactive controller for audio object localization for an arbitrary stereo mixed source.
This product are based on the audio coding framework enabling the extraction and localization operation of audio objects (e.g., vocal, guitar, drums) via the temporal quantization of spatial information [1].
In the last decade, the reproduction of music with a high sense of reality has been superseded with the development of sound reproduction systems with a creative sense of reality in which a user can operate each sound object individually in the audio signal.
Therefore, we proposed an audio object controller developed for interactive sound field reproduction using audio object localization based on spatial representative vector operations.
This system allows a listener to operate an initial sound field and build a new virtual sound field by performing sound field operations on a stereo mixed source.
[1] S. Miyabe, K. Masatoki, H. Saruwatari, K. Shikano, T. Nomura, “Temporal quantization of spatial information using directional clustering for multichannel audio coding,” Proc. of WASPAA, pp.261-264, 2009.
[2] N. Kamado, H. Nawata, H. Saruwatari, K. Shikano, T. Nomura, “Interactive controller for audio object localization based on spatial representative vector operation,” Proc. of IWAENC, 2010.
This figure shows the controller which is equipped with a capacitive touchscreen panel, and the listener can intuitively operate every audio object displayed on the touchscreen panel with his/her finger or a touch pen.
The results of the subjective assessment clarified that this controller enables the listener to change the localization of audio objects without sound degradation if the gain operation is not extreme [2].
Copyright (C) 2011 Speech And Acoustics Laboratory in NAIST. All Rights Reserved.
Last update: 2011.12.27