TY - GEN
T1 - Accurate estimation of tone duration in Automatic music transcription
AU - Funakoshi, Noboru
AU - Tanaka, Toshiyuki
PY - 2010/1/1
Y1 - 2010/1/1
N2 - Although the recent research in the field of Automatic music transcription has been active, there are still outstanding problems. Amongst, we focused on note-length and aim to estimate note-length exactly. In the past, Fourier or Wavelet transform are used to transform music signal into frequency domain. However, these transform can lose accuracy in time component. In this paper, we estimate tone duration from music signal without transforming it into frequency domain.
AB - Although the recent research in the field of Automatic music transcription has been active, there are still outstanding problems. Amongst, we focused on note-length and aim to estimate note-length exactly. In the past, Fourier or Wavelet transform are used to transform music signal into frequency domain. However, these transform can lose accuracy in time component. In this paper, we estimate tone duration from music signal without transforming it into frequency domain.
KW - Automatic music transcription
KW - Correlation
KW - Music note-length extract
KW - Short-time Fourier transform
UR - http://www.scopus.com/inward/record.url?scp=78649279788&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=78649279788&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:78649279788
SN - 9784907764364
T3 - Proceedings of the SICE Annual Conference
SP - 588
EP - 590
BT - Proceedings of SICE Annual Conference 2010, SICE 2010 - Final Program and Papers
PB - Society of Instrument and Control Engineers (SICE)
ER -