Analysis and Prediction of the Audio Feature Space when Mixing Raw Recordings into Individual Stems
×
Cite This
Citation & Abstract
MA. A.. Martinez Ramirez, and JO. D.. Reiss, "Analysis and Prediction of the Audio Feature Space when Mixing Raw Recordings into Individual Stems," Paper 9848, (2017 October.). doi:
MA. A.. Martinez Ramirez, and JO. D.. Reiss, "Analysis and Prediction of the Audio Feature Space when Mixing Raw Recordings into Individual Stems," Paper 9848, (2017 October.). doi:
Abstract: Processing individual stems from raw recordings is one of the first steps of multitrack audio mixing. In this work we explore which set of low-level audio features are sufficient to design a prediction model for this transformation. We extract a large set of audio features from bass, guitar, vocal, and keys raw recordings and stems. We show that a procedure based on random forests classifiers can lead us to reduce significantly the number of features and we use the selected audio features to train various multi-output regression models. Thus, we investigate stem processing as a content-based transformation, where the inherent content of raw recordings leads us to predict the change of feature values that occurred within the transformation.
@article{martinez ramirez2017analysis,
author={martinez ramirez, marco a. and reiss, joshua d.},
journal={journal of the audio engineering society},
title={analysis and prediction of the audio feature space when mixing raw recordings into individual stems},
year={2017},
volume={},
number={},
pages={},
doi={},
month={october},}
@article{martinez ramirez2017analysis,
author={martinez ramirez, marco a. and reiss, joshua d.},
journal={journal of the audio engineering society},
title={analysis and prediction of the audio feature space when mixing raw recordings into individual stems},
year={2017},
volume={},
number={},
pages={},
doi={},
month={october},
abstract={processing individual stems from raw recordings is one of the first steps of multitrack audio mixing. in this work we explore which set of low-level audio features are sufficient to design a prediction model for this transformation. we extract a large set of audio features from bass, guitar, vocal, and keys raw recordings and stems. we show that a procedure based on random forests classifiers can lead us to reduce significantly the number of features and we use the selected audio features to train various multi-output regression models. thus, we investigate stem processing as a content-based transformation, where the inherent content of raw recordings leads us to predict the change of feature values that occurred within the transformation.},}
TY - paper
TI - Analysis and Prediction of the Audio Feature Space when Mixing Raw Recordings into Individual Stems
SP -
EP -
AU - Martinez Ramirez, Marco A.
AU - Reiss, Joshua D.
PY - 2017
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - October 2017
TY - paper
TI - Analysis and Prediction of the Audio Feature Space when Mixing Raw Recordings into Individual Stems
SP -
EP -
AU - Martinez Ramirez, Marco A.
AU - Reiss, Joshua D.
PY - 2017
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - October 2017
AB - Processing individual stems from raw recordings is one of the first steps of multitrack audio mixing. In this work we explore which set of low-level audio features are sufficient to design a prediction model for this transformation. We extract a large set of audio features from bass, guitar, vocal, and keys raw recordings and stems. We show that a procedure based on random forests classifiers can lead us to reduce significantly the number of features and we use the selected audio features to train various multi-output regression models. Thus, we investigate stem processing as a content-based transformation, where the inherent content of raw recordings leads us to predict the change of feature values that occurred within the transformation.
Processing individual stems from raw recordings is one of the first steps of multitrack audio mixing. In this work we explore which set of low-level audio features are sufficient to design a prediction model for this transformation. We extract a large set of audio features from bass, guitar, vocal, and keys raw recordings and stems. We show that a procedure based on random forests classifiers can lead us to reduce significantly the number of features and we use the selected audio features to train various multi-output regression models. Thus, we investigate stem processing as a content-based transformation, where the inherent content of raw recordings leads us to predict the change of feature values that occurred within the transformation.
Authors:
Martinez Ramirez, Marco A.; Reiss, Joshua D.
Affiliation:
Queen Mary University of London, London, UK
AES Convention:
143 (October 2017)
Paper Number:
9848
Publication Date:
October 8, 2017Import into BibTeX
Subject:
Recording and Production
Permalink:
http://www.aes.org/e-lib/browse.cfm?elib=19245