Etienne Parizet

Environmental Sound Perception: Metadescription and Modeling Based on Independent Primary Studies

The aim of the study is to transpose and extend to a set of environmental sounds the notion of sound descriptors usually used for musical sounds. Four separate primary studies dealing with interior car sounds, air-conditioning units, car horns, and closing car doors are considered collectively. The corpus formed by these initial stimuli is submitted to new experimental studies and analyses, both for revealing metacategories and for defining more precisely the limits of each of the resulting categories. In a second step, the new structure is modeled: common and specific dimensions within each category are derived from the initial results and new investigations of audio features are performed. Furthermore, an automatic classifier based on two audio descriptors and a multinomial logistic regression procedure is implemented and validated with the corpus.

Noise assessment in a high-speed train

Binaural sound recordings in various positions in a high-speed train have been used as stimuli in listening tests. In a first experiment, sounds were submitted to subjects at their real levels. Loudness revealed to be the most important parameter; it could be very well described by A-weighted levels. In a second experiment, loudness of stimuli were equalised; strong inter-individual differences appeared, which are due to different appreciations of the low-frequency content of sounds. Usual physical or psycho-acoustical parameters failed to correctly described that perceptual parameter; a simple parameter based on the specific loudness calculations was developed and could correctly describe the results.

Multi-dimensional listening test: selection of sound descriptors and design of the experiment

A method for the selection of parameters used in multi-dimensional listening tests is presented. It allows the presentation to subjects, through a very short learning process, of a complete set of parameters which are unambiguously understood. This set is built after a first test in which other listeners are asked to freely describe differences and similarities between sounds. A method of analysing these free verbalizations allows the determination of peculiarities for sounds belonging to the same context. Two multi-dimensional test methods are compared. The method in which all sounds (or pairs of sounds) are evaluated according to each parameter give slightly more reliable results than the classical method, in which each sound or pair of sounds is evaluated according to all parameters before repeating the operation for the next pair or sounds. These two methods are examined using the idling noise of small diesel cars.

Investigation on localisation accuracy for first and higher order ambisonics reproduced sound sources

Ambisonics and higher order ambisonics (HOA) technologies aim at reproducing sound field either synthesised or previously recorded with dedicated microphones. Based on a spherical harmonic decomposition, the sound field is more precisely described when higher-order components are used. The presented study evaluated the perceptual and objective localisation accuracy of the sound field encoded with four microphones of order one to four and decoded over a ring of loudspeakers. A perceptual test showed an improvement of the localisation with higher order ambisonic microphones. Reproduced localisation indices were estimated for the four microphones and the respective synthetic systems of order one to four. The perceptual and objective analysis revealed the same conclusions. The localisation accuracy depends on the ambisonic order as well as the source incidence. Furthermore, impairments linked to the microphones were highlighted.

Perception and Annoyance Due to Vibrations in Dwellings Generated From Ground Transportation: A Review

Over the past decades, the improvement of the quality of buildings regarding sound insulation has made it possible to reduce noise exposure. But this improvement has led to the increasing importance of other environmental nuisances such as vibrations. Just like noise, vibrations can also affect human health and well-being. National authorities are progressively becoming aware of the adverse effects of vibrations on human beings. One of these adverse effects is vibration annoyance. As a prerequisite for studying vibration annoyance, many research works have provided knowledge of how vibrations are perceived by human beings. But, until now, only a few large-scale studies have been carried out to gain knowledge of annoyance due to vibrations. This knowledge is of importance for public authorities in order to plan the development of transportation networks while taking account of this environmental nuisance (i.e. vibrations). Assessing vibration annoyance is a complex task that implies research work spread over various disciplines (e.g. human sciences, physics). This is why a review of knowledge in the field of vibration perception and annoyance is felt to be necessary. More specifically, this review focuses on vibration annoyance felt by inhabitants inside buildings and caused by ground transportation sources. Before dealing with the review of annoyance due to vibrations, the basics of human vibratory perception are presented. Then, annoyance due to a vibration exposure is dealt with. In particular, the way vibration exposure and vibration annoyance are measured and assessed is tackled. The main features and results of research works that aimed to establish exposure-effect relationships between annoyance and vibrations are described. Finally, this review presents a state-of-the-art of annoyance due to a combined exposure to vibrations and noise, since people exposed to vibrations are also often exposed to airborne noise. This review is non-exhaustive, but it can give to scientists and engineers working in the field tools for a better understanding of some of the factors to consider when dealing with the issue of vibration annoyance due to ground transportation.

Vibro-acoustical comfort in cars at idle: human perception of simulated sounds and vibrations from 3- and 4-cylinder diesel engines

A bench was used to reproduce the vertical vibrations of the seat, the vibrations of the steering wheel and the noise measured at the drivers ears of cars running at hot idle. In a first experiment, subjects were only submitted to seat vertical vibrations; it appeared that comfort could be described by rms values of signals, without using ISO-2631 weightings. In a second experiment, subjects were submitted to all stimuli; the analysis of free verbalisations could explain the perceptual dimensions and were helpful for identifying the contribution of noise and vibration stimuli to overall comfort.

Detectability and annoyance of warning sounds for electric vehicles

Electric or hybrid vehicles are very quiet at low speeds, which represents a very good opportunity to reduce traffic noise annoyance in cities. On the other hand, this may be very hazardous for vulnerable pedestrians (e.g. visually impaired people). The aim of the eVADER project is to propose solutions in order to add warning sounds to such cars, while fulfilling two contradictory goals : sounds should be detectable but should not contribute to traffic noise annoyance. Different perceptual experiments have been conducted : the first one evaluated the influence of various timbre parameters on sound detectability. It was shown that an electric vehicle equipped with one particular sound was as easily detected as a diesel one, while keeping a very low level. Then the influence of some timbre parameters (pitch and temporal modulation frequency) on the distance and speed as perceived by listeners was measured. Finally, a third experiment was devoted to the consequence on traffic noise annoyance of such warning sounds.

Influence of informational content of background noise on speech quality evaluation for VoIP application

With the rise of mobile communication systems, the background noise in the speakers environment and its interaction with VoIP network impairment affects speech quality perception. This effect should be taken into account in nonintrusive models in order to improve accuracy of end user perception measurement. The aim of this study is to determine the impact of information contained in background noise (background speech, environmental sources) on speech quality perception. A subjective test on the speech quality perception in real network conditions has been done comparing the effect of stationary background noise mainly due to transmission equipment (electrical humming or blowing noises) with the effect of nonstationary environmental noise (public place, traffic noise, background conversation). Interactions between these different background noise condition and realistic network impairments (coders and packets loss) were also studied. The speech quality was evaluated through the Mean Opinion Score determi...

Work performance and mental workload in multiple talker environments

The impairment of cognitive performance resulting from the presence of speech sounds is known to increase as the intelligibility of the speech signals is improved. For that reason, speech intelligibility measures are used to quantify the nuisance potential of an unattended voice. However, most of these indexes struggle with situations in which the level of the masking sound is fluctuating. This is the case in open-plan offices, where competing voices are involved. This paper relates a set of experiments in which subjects had to carry out a basic memory task in various office-like noise settings. In addition to a target speech, the masking sounds were made up of speech and differed in level or temporal variability. Disturbance was assessed both through objective measurements of performance and subjective reports of workload. The results highlight the importance of taking into account the temporal fluctuations of the overall ambient sound when trying to ascertain the influence of speech intelligibility on observed and perceived disturbance during the performing of a mental activity. Insights are provided which could lead to the use of a speech intelligibility measure better equipped to deal with multi-sources environments.

Investigation on the restitution system influence over perceived Higher Order Ambisonics sound field: a subjective evaluation involving from first to fourth order systems

Among the spatial audio reproduction techniques over loudspeakers, the Higher Order Ambisonics (HOA) approach is based on a sound field spherical harmonics decomposition. By truncating the decomposition to the Mth order, it remains a finite number of components that form the spatial HOA format. The more components are used to encode the sound field, the finer the spatial resolution is. Similarly, the size of the area where the sound field is accurately recreated is proportional to the order. For an Mth encoding order, N=2M+2 equally distributed loudspeakers are recommended for a homogeneous reproduction in the horizontal plane. Adding loudspeakers does not change the spatial resolution. However, what is the influence of the restitution system on the perceived sound field? An experiment was designed in order to compare four systems (from first to fourth order) and a reference one, using similarity ratings obtained from pairwise comparisons. Two sound scenes were used, simulating an audio conference and a s...