E. T. Chourdakis and J. D. Reiss, “Tagging and Retrieval of Room Impulse Responses Using Semantic Word Vectors and Perceptual Measures of Reverberation,” in Proc. AES Convention 146, Mar. 2019, pp. 10198:1-11. [Online]. Available: https://aes.org/publications/elibrary-page/?id=20331
Chourdakis ET, Reiss JD. Tagging and Retrieval of Room Impulse Responses Using Semantic Word Vectors and Perceptual Measures of Reverberation. In: AES Convention 146. Audio Engineering Society; 2019.pp. 10198:1-11. Available from: https://aes.org/publications/elibrary-page/?id=20331
@inproceedings{Chourdakis2019_20331,
author = {Chourdakis, Emmanouil Theofanis and Reiss, Joshua D.},
title = {{Tagging and Retrieval of Room Impulse Responses Using Semantic Word Vectors and Perceptual Measures of Reverberation}},
booktitle = {AES Convention 146},
pages = {10198:1-11},
url = {https://aes.org/publications/elibrary-page/?id=20331},
year = {2019},
month = mar,
publisher = {Audio Engineering Society},
address = {Dublin Ireland}
}
TY - CPAPER
TI - Tagging and Retrieval of Room Impulse Responses Using Semantic Word Vectors and Perceptual Measures of Reverberation
AU - Chourdakis, Emmanouil Theofanis
AU - Reiss, Joshua D.
T2 - AES Convention 146
SP - 1
EP - 11
M1 - Paper 10198
PY - 2019
DA - 2019/03/06
UR - https://aes.org/publications/elibrary-page/?id=20331
PB - Audio Engineering Society
CY - Dublin Ireland
LA - en
AB - This paper studies tagging and retrieval of room impulse responses from a labelled library. A similarity-based method is introduced that relies on perceptually relevant characteristics of reverberation. This method is developed using a publicly available dataset of algorithmic reverberation settings. Semantic word vectors are introduced to exploit semantic correlation among tags and allow for unseen words to be used for retrieval. Average precision is reported on a subset of the dataset as well as tagging of recorded room impulse responses. The developed approach manages to assign downloaded room impulse responses to tags that match their short descriptions. Furthermore, introducing semantic word vectors allows it to perform well even when large portions of the training data have been replaced by synonyms.
ER -