東華大學圖書館 |

Perceptual and Quantitative Comparison of Headphone-Based Spatial Audio Techniques With Applications in Spatial Enhancement of Music and Binaural Panning /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Perceptual and Quantitative Comparison of Headphone-Based Spatial Audio Techniques With Applications in Spatial Enhancement of Music and Binaural Panning // Zane T Rusk.
Author:	Rusk, Zane T.,
Description:	1 electronic resource (410 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
Contained By:	Dissertations Abstracts International86-03B.
Subject:	User interface. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31353861
ISBN:	9798384212041

Perceptual and Quantitative Comparison of Headphone-Based Spatial Audio Techniques With Applications in Spatial Enhancement of Music and Binaural Panning /
Rusk, Zane T.,

Perceptual and Quantitative Comparison of Headphone-Based Spatial Audio Techniques With Applications in Spatial Enhancement of Music and Binaural Panning /Zane T Rusk. - 1 electronic resource (410 pages)

Source: Dissertations Abstracts International, Volume: 86-03, Section: B.

Headphone-based spatial audio finds many consumer, industrial, and academic applications - anywhere a realistic sound field needs to be presented to a human listener over headphones. However, the nontriviality of the challenges associated with the faithful reproduction of spatial auditory cues leaves many research questions yet unanswered. Broadly speaking, the goal of the current work was to make contributions to the understanding, engineering, and application of headphone-based spatial audio techniques.First, work was focused on a specific application area of binaural audio. Three methods for the spatial enhancement of stereophonic music over headphones were proposed and evaluated in a listening study. Two of these essentially replicated the spatial sound field arising from two loudspeakers and a subwoofer playing in a living room, while a third added no reverberation. Multiple music excerpts and options for spatial resolution were tested for each of the three methods, and four perceptual attributes were rated (Naturalness, Envelopment, Bass-Treble Balance, and Preference). Results indicated that the addition of spatial reverberation was generally not found to be more natural nor preferred to the no-reverberation option. Perceptions of envelopment were increased by the addition of spatial reverberation regardless of spatial resolution. Lastly, the choice of musical excerpt modulated the Naturalness and Envelopmentratings for each enhancement method, but only impacted the Preference and Bass-Treble Balance ratings of the reverberant enhancements.Fundamental aspects of binaural techniques were also investigated in this work. The second study compared several methods of rendering individual sound sources, called binaural panning techniques. These methods were used to pan free-field sound sources, which is analogous to the "reconstruction" of a head-related impulse response (HRIR). The studied techniques included higher-order Ambisonics (HOA) panning (Zotter and Frank 2019), vector-base amplitude panning (VBAP; Pulkki 1997), least-squares (Rafaely and Avni 2010) and Magnitude Least-Squares (MagLS; Schorkhuber et al. 2018) methods of binaural rendering, and the recently-proposed principal component-base amplitude panning (PCBAP; Neal and Zahorik 2022). Overall perceptual similarity between the original HRIRs and reconstructions was evaluated in a listening test. It was confirmed that time-aligning the HRIRs prior to panning is effective in reducing the number of filters required to achieve high fidelity. PCBAP, which is based on principal components analysis of the HRIR set, was also found to be particularly efficient at retaining fidelity.Models for the prediction of perceptual fidelity of binaural panning techniques using quantitative metrics were also generated, complementing the listening test data from the second study. Eight metrics were calculated for each HRIR reconstruction with respect to the original HRIRs. These metrics included the difference in interaural time delay (ITD), interaural level difference (ILD), and modeled binaural loudness. Spectral error metrics were also used, including Perceptual Spectral Difference (PSD) for each ear (Armstrong et al. 2018), normalized mean square error (NMSE) for each ear, and difference in the energetic HRTF magnitude sum across ears. The error metric values for every HRIR reconstruction were used to develop logistic regression models. These models, when assessing each of the two tested source directions individually, were shown to be quite accurate in identifying whether or not an HRIR reconstruction received a high rating for perceptual similarity to the original in the second listening study.

English

ISBN: 9798384212041Subjects--Topical Terms:

3681528
User interface.

Perceptual and Quantitative Comparison of Headphone-Based Spatial Audio Techniques With Applications in Spatial Enhancement of Music and Binaural Panning /
LDR:05228nmm a22003973i 4500 001 2436866
005 20251117053640.5
006 m o d
007 cr|nu||||||||
008 261011s2024 miu||||||m |||||||eng d
020 $a 9798384212041
035 $a (MiAaPQD)AAI31353861
035 $a (MiAaPQD)PennState20721ztr4
035 $a AAI31353861
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Rusk, Zane T., $e author. $3 3827677
245 1 0 $a Perceptual and Quantitative Comparison of Headphone-Based Spatial Audio Techniques With Applications in Spatial Enhancement of Music and Binaural Panning / $c Zane T Rusk.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (410 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
500 $a Advisors: Vigeant-Haas, Michelle C.; Sparrow, Victor W. Committee members: Freihaut, James; Durmus, Dorukalp; Messner, John; Roths, Scott.
502 $b Ph.D. $c The Pennsylvania State University $d 2024.
520 $a Headphone-based spatial audio finds many consumer, industrial, and academic applications - anywhere a realistic sound field needs to be presented to a human listener over headphones. However, the nontriviality of the challenges associated with the faithful reproduction of spatial auditory cues leaves many research questions yet unanswered. Broadly speaking, the goal of the current work was to make contributions to the understanding, engineering, and application of headphone-based spatial audio techniques.First, work was focused on a specific application area of binaural audio. Three methods for the spatial enhancement of stereophonic music over headphones were proposed and evaluated in a listening study. Two of these essentially replicated the spatial sound field arising from two loudspeakers and a subwoofer playing in a living room, while a third added no reverberation. Multiple music excerpts and options for spatial resolution were tested for each of the three methods, and four perceptual attributes were rated (Naturalness, Envelopment, Bass-Treble Balance, and Preference). Results indicated that the addition of spatial reverberation was generally not found to be more natural nor preferred to the no-reverberation option. Perceptions of envelopment were increased by the addition of spatial reverberation regardless of spatial resolution. Lastly, the choice of musical excerpt modulated the Naturalness and Envelopmentratings for each enhancement method, but only impacted the Preference and Bass-Treble Balance ratings of the reverberant enhancements.Fundamental aspects of binaural techniques were also investigated in this work. The second study compared several methods of rendering individual sound sources, called binaural panning techniques. These methods were used to pan free-field sound sources, which is analogous to the "reconstruction" of a head-related impulse response (HRIR). The studied techniques included higher-order Ambisonics (HOA) panning (Zotter and Frank 2019), vector-base amplitude panning (VBAP; Pulkki 1997), least-squares (Rafaely and Avni 2010) and Magnitude Least-Squares (MagLS; Schorkhuber et al. 2018) methods of binaural rendering, and the recently-proposed principal component-base amplitude panning (PCBAP; Neal and Zahorik 2022). Overall perceptual similarity between the original HRIRs and reconstructions was evaluated in a listening test. It was confirmed that time-aligning the HRIRs prior to panning is effective in reducing the number of filters required to achieve high fidelity. PCBAP, which is based on principal components analysis of the HRIR set, was also found to be particularly efficient at retaining fidelity.Models for the prediction of perceptual fidelity of binaural panning techniques using quantitative metrics were also generated, complementing the listening test data from the second study. Eight metrics were calculated for each HRIR reconstruction with respect to the original HRIRs. These metrics included the difference in interaural time delay (ITD), interaural level difference (ILD), and modeled binaural loudness. Spectral error metrics were also used, including Perceptual Spectral Difference (PSD) for each ear (Armstrong et al. 2018), normalized mean square error (NMSE) for each ear, and difference in the energetic HRTF magnitude sum across ears. The error metric values for every HRIR reconstruction were used to develop logistic regression models. These models, when assessing each of the two tested source directions individually, were shown to be quite accurate in identifying whether or not an HRIR reconstruction received a high rating for perceptual similarity to the original in the second listening study.
546 $a English
590 $a School code: 0176
650 4 $a User interface. $3 3681528
650 4 $a Music. $3 516178
650 4 $a Fourier transforms. $3 3545926
650 4 $a Microphones. $3 1458596
650 4 $a Labeling. $3 3560710
650 4 $a Plot (Narrative). $3 3563398
650 4 $a Adaptation. $3 3562958
650 4 $a Data processing. $3 680224
650 4 $a Listening. $3 530843
650 4 $a Energy. $3 876794
650 4 $a Visualization. $3 586179
650 4 $a Sound. $3 542298
650 4 $a Mathematics. $3 515831
690 $a 0791
690 $a 0413
690 $a 0405
710 2 $a The Pennsylvania State University. $e degree granting institution. $3 3765934
720 1 $a Vigeant-Haas, Michelle C. $e degree supervisor.
720 1 $a Sparrow, Victor W. $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-03B.
790 $a 0176
791 $a Ph.D.
792 $a 2024
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31353861