AES E-Library

SU. G.. Bharitkar, T. Mauer, T. Wells, and D. Berfanger, "Bayesian Optimization of Deep Learning Techniques for Synthesis of Head-Related Transfer Functions," Paper 10162, (2019 March.). doi: SU. G.. Bharitkar, T. Mauer, T. Wells, and D. Berfanger, "Bayesian Optimization of Deep Learning Techniques for Synthesis of Head-Related Transfer Functions," Paper 10162, (2019 March.). doi:
Abstract: Head-related transfer functions (HRTF) are used for creating the perception of a virtual sound source at horizontal angle ø and vertical angle ?. Publicly available databases use a subset of a full-grid of angular directions due to time and complexity to acquire and deconvolve responses. In this paper we build up on our prior research [5] by extending the technique to HRTF synthesis, using the IRCAM dataset, while reducing the computational complexity of the autoencoder (AE)+fully-connected-neural-network (FCNN) architecture by ˜ 60% using Bayesian optimization. We also present listening test results, demonstrating the performance of the presented approach, from a pilot study that was designed for assessing the directional cues of the proposed architecture.

@article{bharitkar2019bayesian,
author={bharitkar, sunil g. and mauer, timothy and wells, teresa and berfanger, david},
journal={journal of the audio engineering society},
title={bayesian optimization of deep learning techniques for synthesis of head-related transfer functions},
year={2019},
volume={},
number={},
pages={},
doi={},
month={march},} @article{bharitkar2019bayesian,
author={bharitkar, sunil g. and mauer, timothy and wells, teresa and berfanger, david},
journal={journal of the audio engineering society},
title={bayesian optimization of deep learning techniques for synthesis of head-related transfer functions},
year={2019},
volume={},
number={},
pages={},
doi={},
month={march},
abstract={head-related transfer functions (hrtf) are used for creating the perception of a virtual sound source at horizontal angle ø and vertical angle ?. publicly available databases use a subset of a full-grid of angular directions due to time and complexity to acquire and deconvolve responses. in this paper we build up on our prior research [5] by extending the technique to hrtf synthesis, using the ircam dataset, while reducing the computational complexity of the autoencoder (ae)+fully-connected-neural-network (fcnn) architecture by ˜ 60% using bayesian optimization. we also present listening test results, demonstrating the performance of the presented approach, from a pilot study that was designed for assessing the directional cues of the proposed architecture.},}

TY - paper
TI - Bayesian Optimization of Deep Learning Techniques for Synthesis of Head-Related Transfer Functions
SP - EP -
AU - Bharitkar, Sunil G.
AU - Mauer, Timothy
AU - Wells, Teresa
AU - Berfanger, David
PY - 2019
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - March 2019 TY - paper
TI - Bayesian Optimization of Deep Learning Techniques for Synthesis of Head-Related Transfer Functions
SP - EP -
AU - Bharitkar, Sunil G.
AU - Mauer, Timothy
AU - Wells, Teresa
AU - Berfanger, David
PY - 2019
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - March 2019
AB - Head-related transfer functions (HRTF) are used for creating the perception of a virtual sound source at horizontal angle ø and vertical angle ?. Publicly available databases use a subset of a full-grid of angular directions due to time and complexity to acquire and deconvolve responses. In this paper we build up on our prior research [5] by extending the technique to HRTF synthesis, using the IRCAM dataset, while reducing the computational complexity of the autoencoder (AE)+fully-connected-neural-network (FCNN) architecture by ˜ 60% using Bayesian optimization. We also present listening test results, demonstrating the performance of the presented approach, from a pilot study that was designed for assessing the directional cues of the proposed architecture.