Spectral transformation for musical tones via time domain filtering

We present a spectral transformation technique for musical tones. It can be used to modify the brightness or timbre of a musical tone. We perform spectral transformation by modifying the line spectral frequencies or LPC roots of the original spectral envelope and filtering the tone with a spectral transformation filter in the time domain. One of the application is pitch modification where frequency scaling is used to modify the fundamental frequency of a tone and spectral transformation is used to capture the original brightness and timbre. We have applied this spectral transformation technique to restore the original brightness of frequency scaled musical tones produced from a variety of musical instruments, such as piano, trumpet, violin, flute and bassoon by one octave up or down. In all experiments conducted, the pitch modified tones resemble the original brightness and timbre. This technique allows us to control the timbre of musical tones such as brightness, friction or other modulation by manipulating the spectral parameters.

[1]  Eric Moulines,et al.  HNS: Speech modification based on a harmonic+noise model , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  J. Risset,et al.  Exploration of timbre by analysis and synthesis , 1999 .

[3]  Thomas F. Quatieri,et al.  Speech transformations based on a sinusoidal representation , 1986, IEEE Trans. Acoust. Speech Signal Process..

[4]  Yinong Ding,et al.  Processing of Musical Tones Using a Combined Quadratic Polynomial-Phase Sinusoid and Residual (QUASAR) Signal Model , 1997 .

[5]  Eric Moulines,et al.  Pitch-synchronous waveform processing techniques for text-to-speech synthesis using diphones , 1989, Speech Commun..

[6]  A. El-Jaroudi,et al.  Time-scale modification in medium to low rate speech coding , 1986, ICASSP '86. IEEE International Conference on Acoustics, Speech, and Signal Processing.