Code
Audio Time-Scale Modification Algorithms
- hybrid.m A hybrid time-scale modification algorithm based upon A Hybrid Time-Frequency Domain Approach to Audio Time-Scale Modification, Dorran, David; Lawlor, Bob; Coyle, Eugene, Journal of The Audio Engineering Society 2006. Offers an improvement for time-scaling speech over phase locked vocoder approach, but no improvement for music. Requires a zeros_crossings function
- pl_phaseVocoder_variable_analysis_hop.m and pl_phaseVocoder_fixed_analysis_hop.m Phase locked vocoder time-scale modification algorithms based upon Jean Laroche's 1999 work (Improved Phase Vocoder, IEEE Trans on speech and audio processing). The fixed analysis hop version incorporates Jordi Bonada's (2000) approach . This code started out as the code provided by Tae Hong Park, but has changed significantly over the years.
- vsola.m This time-domain implementation makes use of the standard SOLA algorithm (Rocus and Wilgus, ICASSP 1986) and some efficient paramter settings for the SOLA algorithm (Dorran, Lawlor and Coyle, ICASSP 2003) and (Dorran, Lawlor and Coyle, DAFX 2003). Requires a linear_cross_fade function
- tsm_mstft.m An implementation based on the magnitude only reconstruction algorithm of Griffen and Lim (IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-32(2), pp.236-243, April 1984.)