Listening Samples

Title: Optimizations of Neural Audio Coder Toward Perceptual Transparency (submitted to IEEE JSTSP Special Issue on Neural Speech and Audio Coding)

This page presents audio clips for comparative listening between our neural audio coder (NAC) and the commercial audio coder (AAC-LC with Fraunhofer FDK AAC encoder), operating at bitrates from 48kbps to 64kbps.

** Please download audio clips from the link below.
(https://drive.google.com/file/d/1R-0uNqmQFQbm7FscLleV2q-9m90mmbl3/view?usp=drive_link)

References

[38] S. Shin, J. Byun, Y. Park, J. Sung, and S. Beack, “Deep neural network (DNN) audio coder using a perceptually improved training method,” in Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2022, pp. 871–875. (DOI: 10.1109/ICASSP43922.2022.9747575)

[39] 2. J. Byun, S. Shin, Y. Park, J. Sung, and S. Beack, “A perceptual neural audio coder with a mean-scale hyperprior,” in Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023, pp. 1–5. (DOI: 10.1109/ICASSP49357.2023.10096009)