Final Year Project 2020
An Investigation of the Coherent-to-Diffuse Power Ratio and its Application to Speech Dereverberation
This repository contains all the code scripts used for the project, including
- Room Impulse Response (RIR) examples
- CDR investigations (adapted from the demo code by Schwarz et al.)
- CDR applied to dereverberation and evaluation
The scripts uploaded are only for academic purposes. Any external dependencies or helper functions used are stored in the 'resources' folder; they are not uploaded (to avoid plagiarism) but can be found via the references. References are included in individual scripts as well as the project final report.
References
[1] A. Schwarz and W. Kellermann, “Coherent-to-diffuse power ratio estimation for dere- verberation,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 23, pp. 1006–1018, June 2015. (https://github.com/andreas12345/cdr-dereverb)
[2] P. Naylor and N. Gaubitch, Speech Dereverberation. Springer, 2010.
[3] T. Nakatani, T. Yoshioka, K. Kinoshita, M. Miyoshi, and B. Juang, “Speech derever- beration based on variance-normalized delayed linear prediction,” IEEE Transactions on Audio, Speech, and Language Processing, vol. 18, no. 7, pp. 1717–1731, 2010. (http://www.kecl.ntt.co.jp/icl/signal/wpe/)
[4] ludlows, “The MEX wrapper for PESQ (perceptual evaluation of speech quality)” (https://github.com/ludlows/pesq-mex).
[5] J. Eaton, N. D. Gaubitch, A. H. Moore, and P. A. Naylor, “Estimation of room acoustic parameters: The ACE challenge,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 24, pp. 1681–1693, Oct 2016.
[6] E. Habets, “RIR-generator” (https://github.com/ehabets/RIR-Generator)