Subjective Evaluation Experiments Research Articles

Listening as your neighbors: Simulation and evaluation of building structural transmission

The enjoyment of music can be jeopardized by the knockings on the door and your furious neighbor shows up. Building structural transmission passes annoying sound to neighbors even when the audio is at low volume. In this paper, we implement a computational model for the simulation and evaluation of building structural transmission based on field measurements of structural transmission profiles and subjective listening tests. This model is intended to serve as a signal analysis tool that allows the audio engineers to evaluate the interference level using computer software and monitor system, so they can assess and mitigate these interferences without the need of conducting tedious subjective evaluation experiments in real acoustical environments. Our proposed transmission models are summarized from multiple acoustical measurements conducted in real room pairs to ensure its authenticity. Essentially we build “virtual” room pairs so we can simulate the interferences reaching the neighbors. We also implement a computational auditory model based on audio content analysis and subjective evaluation experiments to automatically evaluate the interference level caused by the transmitted sound. The proposed simulation and evaluation tool finds extensive applications in various areas of audio engineering such as audio content production, noise control, and audio system design.

A design of control signal in reducing discomfort of the dental treatment sound based on auditory masking

In dental treatment, patients feel a strong discomfort feeling by the treatment sounds which arise by a tooth grinding. In order to add comfort to quality of life, we aim to reduce the discomfort feeling with dental treatment sounds. We previously proposed the unpleasantness reduction method based on auditory masking to reduce discomfort feeling of noise. The previously proposed method can reduce discomfort feeling by emitting a control signal to a listener, but we had focused on unpleasantness reduction to noise which has a peak frequency. Meanwhile, dental treatment sounds tend to consist of multiple spectral peaks. Therefore, in the present paper, we propose the design method of control signals for reducing discomfort feeling of dental treatment sounds which have multiple spectral peaks. More specifically, we detect the main spectral peaks, which bring a discomfort feeling, and design the control signal, which can mask these spectral peaks. Also, we employ the sound of running water as a source for the control signal. We carried out subjective evaluation experiments to confirm the effectiveness of the proposed method. As a result of evaluation experiments, we confirmed the effectiveness of the proposed method.

An investigation into the relationship between sound quality and demodulation ratio for parametric loudspeakers

Recently, parametric loudspeakers with a powerful ultrasound have been utilized for reproduction of sound to particular area. Amplitude of the ultrasound is modulated with a target audible sound. The emitted amplitude modulated wave is demodulated into the target audible sound in the air. Demodulation ratio affects sound quality of the reproduced sound with parametric loudspeakers. Moreover, demodulation ratio depends on the distance between the parametric loudspeaker and the listener. Therefore, the listener should utilize parametric loudspeakers at the suitable distance, which is required for sufficient demodulation of the amplitude modulated wave. Thus, we have proposed the criterion for measuring the demodulation ratio. However, it had not been conducted to investigate a relationship between subjective sound quality and demodulation ratio. In this paper, we investigate the relationship for estimation of the suitable distance. We carried out subjective evaluation experiment for confirming sound quality at each distance. As a result, we confirmed the consistency about fluctuation tendency of the subjective sound quality and demodulation ratio at each distance.

Steering for listening area of reflective audio spot with parametric loudspeaker array

A parametric loudspeaker has a high directivity by utilizing an ultrasound wave as a carrier wave. Therefore, a parametric loudspeaker can form a specific listening called Furthermore, a parametric loudspeaker can form the audio spot by utilizing the reflected sound. The listeners in the listening area may perceive the acoustic sound image on the reflector. However, it has the problem that the reflective audio with a single parametric loudspeaker is narrow area. Therefore, it is difficult for several listeners to perceive the reflective audio at the same time. If we can expand the area of the reflective audio spot, several listeners can perceive the reflective audio at the same time. Thus in this paper, we attempt to steer the area of the reflective audio with parametric loudspeaker array. We carried out objective and subjective evaluation experiments to confirm the effectiveness of the proposed system in a conference room. As a result, we confirmed the proposed system can expand the are of the reflective audio spot.

Subjective quality assessment of asymmetric stereoscopic 3D video

In asymmetric stereoscopic video compression, the views are coded with different qualities. According to the binocular suppression theory, the perceived quality is closer to that of the higher-fidelity view. Hence, a higher compression ratio is potentially achieved through asymmetric coding. Furthermore, when mixed-resolution coding is applied, the complexity of the coding and decoding is reduced. In this paper, we study whether asymmetric stereoscopic video coding achieves the mentioned claimed benefits. Two sets of systematic subjective quality evaluation experiments are presented in the paper. In the first set of the experiments, we analyze the extent of downsampling for the lower-resolution view in mixed-resolution stereoscopic videos. We show that the lower-resolution view becomes dominant in the subjective quality rating at a certain downsampling ratio, and this is dependent on the sequence, the angular resolution, and the angular width. In the second set of the experiments, we compare symmetric stereoscopic video coding, quality-asymmetric stereoscopic video coding, and mixed-resolution coding subjectively. We show that in many cases, mixed-resolution coding achieves a similar subjective quality to that of symmetric stereoscopic video coding, while the computational complexity is significantly reduced.

Study on measuring the exposure of sound samples in subjective experiments

The traditional measurements for environmental noise exposure-dose are inapplicable to the sound samples with fixed playing duration used in subjective evaluation experiment. In this paper, we propose a method for exposure-dose measurement of sound samples with fixed duration. The method based on the short-term exposure-dose values during the playing time of the sound samples. An entire analysis on the potential influencing factors are carried out, including the short-term analysis period, the statistical properties of the short-term exposure-dose values and four different frequency weightings. Then 11 indices are used to measure the exposure-dose of the fixed duration sound sample, which lays the groundwork for establishing the dose-annoyance relationships via subjective evaluating experiments.

Control Characteristics of Robust Power Assist Cart for Flexible Structures

Power assist systems have been introduced to reduce physical burdens of workers in industrial production lines. The authors have studied on control systems for power assisted conveyance of flexible structures. The previous paper proposed a control system design consisting of an impedance controller, a disturbance accommodating suboptimal controller, a disturbance observer and a reaction force controller in order to achieve the robust assist against structured uncertainties. However, the experimental and simulation consideration in the paper adopted single operational force profile which was reserved a priori and compared control characteristics of plural control system designs. Therefore, responses by actual human operational forces have not been fully evaluated yet. Then, a subjective evaluation experiment is conducted in this paper. As an experimental task, participants push the cart about 10 m distance and position the tip of a flexible plate within 100 mm. The participants answer some questions about the operability of the cart. By using Dunnett's test as a statistical method, the effectiveness of the proposed control system is verified. Moreover, especially at the point of the effectiveness of the disturbance observer, another experiment based on the reserved operational force is conducted on moving on a flat and slope track. By the experiments in this paper in addition to the previous paper results, the control characteristics of the proposed control system design is evaluated comprehensively and multilaterally.

Real and imaginary modulation spectral subtraction for speech enhancement

In this paper, we propose a novel spectral subtraction method for noisy speech enhancement. Instead of taking the conventional approach of carrying out subtraction on the magnitude spectrum in the acoustic frequency domain, we propose to perform subtraction on the real and imaginary spectra separately in the modulation frequency domain, where the method is referred to as MRISS. By doing so, we are able to enhance magnitude as well as phase through spectral subtraction. We conducted objective and subjective evaluation experiments to compare the performance of the proposed MRISS method with three existing methods, including modulation frequency domain magnitude spectral subtraction (MSS), nonlinear spectral subtraction (NSS), and minimum mean square error estimation (MMSE). The objective evaluation used the criteria of segmental signal-to-noise ratio (Segmental SNR), PESQ, and average Itakura–Saito spectral distance (ISD). The subjective evaluation used a mean preference score with 14 participants. Both objective and subjective evaluation results have demonstrated that the proposed method outperformed the three existing speech enhancement methods. A further analysis has shown that the winning performance of the proposed MRISS method comes from improvements in the recovery of both acoustic magnitude and phase spectrum.

Randomized sample-level interpolation for audio content manipulation

In this paper we propose a novel audio signal manipulation algorithm based on sample-level interpolations that can generate multiple unique versions of an audio file without creating any perceptual difference. The proposed algorithm enables important applications such as digital rights management and file distribution tracking. The simplest sample-level interpolation method is based on time domain interpolation of fixed-length audio frames. The processing algorithm first segments the audio signal into fixed-length frames. For each frame, we perform a slight time compression or extension using an audio sample interpolation algorithm and then we recombine the manipulated audio samples to form a manipulated version of the original audio files. To enable better security features a randomization program is applied to control the frame-length and manipulation-length using pseudo-random sequences. The result of this algorithm is effectively a form of weak frequency modulation. If the fame size is larger than the compression/extension sample number, these compression/extension manipulations will not produce any audible difference. Various subjective evaluation experiments are conducted to decide the extent of the admissible processing parameters that will not cause noticeable difference in both fixed-length and randomized-length sample manipulation. The authors also provide several implementation examples and a brief summary.

Associative Motion Generation for Humanoid Robot Reflecting Human Body Movement

This paper proposes an intuitive real-time robot control system using human body movement. Recently, it has been developed that motion generation for humanoid robots with reflecting human body movement, which is measured by a motion capture. However, in the existing studies about robot control system by human body movement, the detailed structure information of a robot, for example, degrees of freedom, the range of motion and forms, must be examined in order to calculate inverse kinematics. In this study, we have proposed Associative Motion Generation as humanoid robot motion generation method which does not need the detailed structure information. The associative motion generation system is composed of two neural networks: nonlinear principal component analysis and Jordan recurrent neural network, and the associative motion is generated with the following three steps. First, the system learns the correspondence relationship between an indication and a motion using training data. Second, associative values are extracted for associating a new motion from an unfamiliar indication using nonlinear principal component analysis. Last, the robot generates a new motion through calculation by Jordan recurrent neural network using the associative values. In this paper, we propose a real-time humanoid robot control system based on Associative Motion Generation, that enables user to control motion intuitively by human body movement. Through the task processing and subjective evaluation experiments, we confirmed the effective usability and affective evaluations of the proposed system.

Three-dimensional acoustic sound field reproduction based on hybrid combination of multiple parametric loudspeakers and electrodynamic subwoofer

Auditory Mixed Reality (MR) systems that reproduce Three-Dimensional (3-D) acoustic sound fields have recently become a research focus because the combination of visual and auditory MR systems can achieve a greater sense of presence than conventional visual MR systems. General auditory MR systems usually use a headphone-based system with a Head-Related Transfer Function (HRTF), which is a major system for reproducing 3-D acoustic sound fields. However, the localization accuracy of sound images with a HRTF depends on the individual. On the other hand, we have already proposed a system for reproducing a 3-D acoustic sound field with parametric loudspeakers instead of headphones. The 3-D acoustic sound field reproduced by this system has achieved a highly accurate localization of sound images. However, one problem is that it is difficult to reproduce lower frequency sounds using parametric loudspeakers, which causes a poorer sound quality. We tried to accomplish a greater sense of presence for 3-D acoustic sound fields based on a hybrid combination of an electrodynamic subwoofer and the parametric loudspeakers by improving the sound quality. Sound images were formed at the target location using the parametric loudspeakers, and a lower frequency sound was compensated for by using the electrodynamic subwoofer. Subjective evaluation experiments were conducted to verify the effectiveness of the proposed system. We confirmed the improved sound quality while maintaining a higher accuracy of sound image localization by using the proposed system. We also confirmed the optimum parameters of the proposed system to achieve a greater sense of presence.

Knowledge Expansion Support by Related Search Keyword Generation Based onWikipedia Category and Pointwise Mutual Information

When users use search engines to acquire knowledge on certain subjects in unknown domains, they often refer to the related search keywords that are generated on the frequency of use as search keywords. However, such searches by reference to related search keywords may not always turn out to be useful for the expansion of knowledge on the research subjects. We, therefore, propose a new method to generate related search keywords by means of Wikipedia. In the proposed method, users first searchWikipedia pages of the same title with the queries input by users to extract information on the category of the pages. Next, obtain the sets of pages that fall into the category and extract related page groups from the pages contained in any plural product sets of pages. Then, calculate pointwise mutual information or tf-idf for the keywords extracted from each page to make either information of higher values associated with search keywords. We have confirmed effectiveness of the proposed method through comparison with related search keywords generated by Google as well as through subjective evaluation experiments.

Kansei Song Selection System Based on Music Fluctuation Features and Comparison of Song Selection Algorithm

Recently, we have so many songs due to a large capacity of storage, however it becomes difficult to select the song using bibliographic data. In this paper, we propose Kansei song selection system based on music fluctuation features which named YSSS. In this system, the acoustic fluctuation features that show time variations of music are extracted from each songs, and the Kansei evaluations of the song are obtained through the subjective evaluation experiments. Then the features and the evaluations are related to each other, and the estimation spaces for Kansei evaluations are constructed. Using this estimation spaces, the Kansei evaluations are labeled with the songs. We proposed two types of song selection method, Exactly-Matching and Tolerant-List which considers the tolerance of human instinct. Through the subjective evaluation experiment, we confirmed the correct song selection and the contentment with proposed system. Furthermore, we confirmed that proposed system with Tolerant-List method shows better usability.

Speech Prior Estimation for Generalized Minimum Mean-Square Error Short-Time Spectral Amplitude Estimator

SUMMARY In this paper, we introduce a generalized minimum meansquare error short-time spectral amplitude estimator with a new prior estimation of the speech probability density function based on momentcumulant transformation. From the objective and subjective evaluation experiments, we show the improved noise reduction performance of the proposed method.

Robust Transmission of Audio Signals over the Internet: An Advanced Packet Loss Concealment for MP3-Based Audio Signals

This paper describes packet loss concealment methods for MP3 audio. The proposed methods are based on estimation of modified discrete cosine transform (MDCT) coefficients of the lost packets. The estimation of MDCT coefficients of lower dimensions is performed by switching two concealment methods: the sign correction method and the correlation-based method. The concealment methods are switched based on redundant side information calculated subband-by-subband for reducing MDCT prediction errors. Next, a method for improving estimation of MDCT coefficients of higher dimensions was proposed. The method estimates the absolute value and sign of an MDCT coefficient independently. The subjective evaluation experiment proved that both of the improvement methods for lower and higher dimensions effectively improved the subjective audio quality.

Handshake: Realistic Human-Robot Interaction in Haptic Enhanced Virtual Reality

This paper focuses on the development and evaluation of a haptic enhanced virtual reality system which allows a human user to make physical handshakes with a virtual partner through a haptic interface. Multimodal feedback signals are designed to generate the illusion that a handshake with a robotic arm is a handshake with another human. Advanced controllers of the haptic interface are developed to respond to user behaviors online. Techniques to achieve online behavior generation are presented, such as a hidden-Markov-model approach to human interaction strategy estimation. Human-robot handshake experiments were carried out to evaluate the performance of the system. Two different approaches to haptic rendering were compared in experiments: a controller in basic mode with an embedded curve in the robot that disregards the human partner, and an interactive robot controller for online behavior generation. The two approaches were compared with the ground truth of another human driving the robot via teleoperation instead of the controller implementing a virtual partner. In the evaluation results, the human approach is rated to be most human-like, with the interactive controller following closely behind, followed by the controller in basic mode. This paper mainly concentrates on discussing the development of the haptic rendering algorithm for the handshaking system, its integration with visual and haptic cues, and reports about the results of subjective evaluation experiments that were carried out.

Subjective evaluation of sound images of air-conditioning system in a vehicle

In this study, we attempted to evaluate air-conditioning sounds in a vehicle from the viewpoint of sound quality. First, we carried out an experiment to select suitable words for evaluating air-conditioning sounds in the compartment of a vehicle. The subjects of the experiment selected several words that were appropriate to represent images of the air-conditioning sounds from a list of adjectives. As a result of a cluster analysis of the experimental data, the seven evaluation words “quiet,” “refreshing,” “heavy,” “wide,” “muddy,” “violent,” and “dry” were selected as the evaluation words. Next, we carried out a subjective evaluation experiment by the rating scale method. The subjects of the experiment evaluated each presented sound in seven degrees for the seven evaluation words. As a result of a factor analysis, the air-conditioning sounds were found to be represented by three factors: “rough,” “space,” and “refreshing.” Finally, we carried out an experiment to evaluate the air-conditioning sounds using psychoacoustic parameters such as loudness and sharpness. As a result, we found that the rough factor correlated with loudness and space and refreshing factors correlated with sharpness.

INCA Algorithm for Training Voice Conversion Systems From Nonparallel Corpora

Most existing voice conversion systems, particularly those based on Gaussian mixture models, require a set of paired acoustic vectors from the source and target speakers to learn their corresponding transformation function. The alignment of phonetically equivalent source and target vectors is not problematic when the training corpus is parallel, which means that both speakers utter the same training sentences. However, in some practical situations, such as cross-lingual voice conversion, it is not possible to obtain such parallel utterances. With an aim towards increasing the versatility of current voice conversion systems, this paper proposes a new iterative alignment method that allows pairing phonetically equivalent acoustic vectors from nonparallel utterances from different speakers, even under cross-lingual conditions. This method is based on existing voice conversion techniques, and it does not require any phonetic or linguistic information. Subjective evaluation experiments show that the performance of the resulting voice conversion system is very similar to that of an equivalent system trained on a parallel corpus.

基于音色属性辨识声目标的主观评价实验研究

Three subjective evaluation experiments involving in target recognition, semantic differential (SD) and adaptive grouped paired comparison (AGPC) were designed and performed to investigate the effect of timbre attributes on acoustic target discrimination. Firstly, 50 sound samples of equalized loudness and duration were used to subjectively identify three kinds of targets. Then SD and AGPC based target discrimination experiments were carried out, and the experimental data were processed using coherence and clustering approach to explore the roles of the two kinds of timbres, natural timbre and essential timbre, in the identification of acoustic targets. This work provides some insight into timbre mechanisms inherent in auditory object identification and timbre modeling applied to auditory feature extraction and target recognition.

Video Quality Assessment Using Spatio-Velocity Contrast Sensitivity Function

Due to the development and popularization of high-definition televisions, digital video cameras, Blu-ray discs, digital broadcasting, IP television and so on, it plays an important role to identify and quantify video quality degradations. In this paper, we propose SV-CIELAB which is an objective video quality assessment (VQA) method using a spatio-velocity contrast sensitivity function (SV-CSF). In SV-CIELAB, motion information in videos is effectively utilized for filtering unnecessary information in the spatial frequency domain. As the filter to apply videos, we used the SV-CSF. It is a modulation transfer function of the human visual system, and consists of the relationship among contrast sensitivities, spatial frequencies and velocities of perceived stimuli. In the filtering process, the SV-CSF cannot be directly applied in the spatial frequency domain because spatial coordinate information is required when using velocity information. For filtering by the SV-CSF, we obtain video frames separated in spatial frequency domain. By using velocity information, the separated frames with limited spatial frequencies are weighted by contrast sensitivities in the SV-CSF model. In SV-CIELAB, the criteria are obtained by calculating image differences between filtered original and distorted videos. For the validation of SV-CIELAB, subjective evaluation experiments were conducted. The subjective experimental results were compared with SV-CIELAB and the conventional VQA methods such as CIELAB color difference, Spatial-CIELAB, signal to noise ratio and so on. From the experimental results, it was shown that SV-CIELAB is a more efficient VQA method than the conventional methods.