The Acoustic and Cultural Heritage of the Banda Primitiva de Llíria Theater: Objective and Subjective Evaluation

Abstract

La Banda Primitiva is one of Europe’s most notable symphonic civic bands. Located in Llíria (València, Spain), part of the UNESCO creative cities network, its theater was designed by Joaquín Rieta, one of the most relevant Valencian architects of the twentieth century. This study analyses the current state of the theater, its relevance to the town’s cultural heritage, and how it has evolved over the years in terms of its acoustic performance. The objective is to understand how the theater’s acoustics have evolved over the years and to unveil the reasons behind the preference of the regular audience for specific areas of the venue, considering if these tendencies are influenced by tradition rather than the current auditory experience. The theater’s acoustics were studied with objective and subjective parameters. The objective parameters were assessed by conducting on-site measurements and ray-tracing simulations. One hundred and three musicians answered a survey of auralizations to evaluate subjective acoustic parameters. Three musical pieces were recorded in an anechoic chamber and convoluted by the impulse response of the venue at different positions to obtain the auralizations. The results show that the objective acoustic parameters do not differ significantly. Overall, the reverberation time was longer before the renovation. Regarding subjective testing, the sample only shows a subtle tendency (57%) towards preferring seats on the second balcony. For that reason, it can be concluded that there is no evidence to support the claim that the seats located on the second balcony have better acoustics than those in the stalls.

1. Introduction

In the Spanish region of València, symphonic music is one of the most recognizable cultural traits. Most towns pride themselves on having at least one high-quality musical ensemble. Those ensembles are structured as music associations that not only focus on playing concerts but also are configured as music schools, where the students eventually become instrumentalists in the symphonic band. These associations are usually deeply rooted in the local society and tradition. They have been vital in the cultural evolution of the region, being involved in the education of the youth and promoting music as a communal leisure activity [1,2].

The most recognizable example of this is the town of Llíria, located in the Camp de Túria region. In 2019, UNESCO accepted Llíria into the creative cities network. This honor solidifies its trajectory in promoting music and culture nationally and internationally [3]. The ties the town has with music are not new. The first evidence of music played in Llíria dates back to the 3rd century BC. Nowadays, Llíria arguably has the two most well-known civic bands in the region and also hosts other musical associations [4,5]. This musical ecosystem constitutes an intangible cultural heritage that has configured life in the town over generations, in which people take pride in being part of a band and helping it grow [6].

Among those bands, the Banda Primitiva de Llíria is one of the country’s most internationally recognized musical associations. The origins of this musical society date back to 1819, that is, it has more than 200 years of history [5,6]. One of its identitying traits is its theater, designed by one of the most notable Valencian architects of the first half of the twentieth century, Joaquín Rieta. Rieta’s better-known projects are the Capitol Cinema, Casa Gil, and Casa Cervera, all built during the thirties in València [7]. However, his architecture after the Spanish Civil War also deserves some attention, with significant projects such as La Salle school in Paterna in 1959. El Teatro de la Primitiva, built in 1951, is among those constructed during the latter part of his career [8]. As discussed in Section 2.1, this theater incorporated a revolutionary structure for the time, allowing for its wide distribution.

Since its construction in 1951, the Banda Primitiva de Llíria theater has hosted many different types of performances [6]. Classical instrumental music concerts like symphonic bands, orchestras, chamber groups, and soloists are the most common. Besides instrumental concerts, there have also been theatrical performances, zarzuelas, operas, acapella music, musicals, and even film screenings. The versatility of its acoustics and its seating capacity of around 1000 people allow this theater to be used as a suitable venue for a wide range of important cultural activities.

Historically, most of the regular attendants of this theater have stated that the best area to listen to and appreciate the sound nuances is the central area of the second balcony. Most of the members and sympathizers of this musical society pass on this tip by word of mouth. However, to this day, there has yet to be any research that corroborates or disproves this statement.

Recognizing this theater’s historical and cultural significance and the music association, this paper investigates its acoustic parameters and their change over time. The objective is to understand how the renovation in 1991, described in the following sections, changed the overall acoustics of the theater and how the auditory experience varies within the attendance area. These effects are studied by combining subjective and objective parameters. The present-day acoustics of the theater were studied through on-site measurements and ray-tracing acoustic simulation. In the case of the original state of the theater, a virtual model was designed to recreate the original materials employed. The information was obtained by accessing the original project plans and documentation. As explained in the methodology section, a sample of professional and amateur musicians conducted a test through auralizations to assess their preferences through subjective parametrization. Auralization is one of the most powerful tools used nowadays for simulating the listening experience of ancient architecture [9].

Researchers have tried to find connections between subjective and objective acoustic parameters for around fifty years. In 1974, a study by researchers from the Drittes Physikalisches Institut of the Universität Göttingen correlated subjective and objective parameters by recording anechoic music with binaural microphones placed inside a dummy head. Those recordings were then provided to musicians, who filled out a survey on the perception of the recorded pieces [10]. Years later, in 1977, a study published by Wilkens investigated that relation by recording an orchestra in six different concert halls and asking musicians to fill out a questionnaire rating the sound characteristics of each fragment [11]. During the following years, researchers continued to work on ways to measure and evaluate the subjective perception of acoustic spaces. This is the case of a study by Mike Barron in 1988 in which musicians and expert listeners evaluated eleven British concert halls. The study found that the subjects studied fell into two clusters: those that preferred ‘Intimacy’ and those that preferred ‘Reverberance’. In 1995, Mike Barron also studied how to combine ‘Intimacy’ and ‘Reverberance’ by altering the Early Decay Time (EDT) and the Reverberation Time (RT) with different geometries and absorbent surfaces [12]. That same year, Sotiropoulou et al. conducted a two-part study evaluating the subjective performance of concert halls. In part one, they rated four parameters, ‘body’, ‘clarity’, ‘tonal quality’, and ‘proximity’, employing bipolar semantic rating scales. In part two, they correlated the subjective parameters with objective ones. They found a good correlation between a low-frequency 80 ms early-to-late sound index and ‘body’ and ‘proximity’ with the total sound level, the early sound level, and the source–receiver distance. They also found that ‘tonal quality’ can be associated with the slope per octave of the EDT. However, the researchers found no relation between ‘clarity’ and their measured physical data [13]. In 2000, Hidaka and Beranek published an article showing the results of a cross-continental study on twenty-three opera houses. The study combined on-site measurements and questionnaires to professionals in the field, mostly opera conductors. That allowed them to find links between the preferences of the conductors and the physical results obtained [14]. Also, in 2000, Semidor and Barlet conducted an objective and subjective analysis of the Grand Theatre of Bordeaux. The study mixed answers from professional musicians and regular attendants, to compare their responses and understand how the preferences of the musicians and the regular audience differ [15]. A year later, in 2001, Angelo Farina published a study analyzing eight Italian theaters and halls and gathering responses on the subjective parameters from almost 200 musicians. The result correlated parameters like EDT with diffuse–concentrated and dry–reverberant, ITDG with soft–hard, and C80 with pleasant–unpleasant. The study also helped translate music terminology into Italian and English [16]. That same year, Tapio Lokki and Hanna Järveläinen developed a survey in which they conducted listening tests with auralizations taken from virtual models. Despite some differences between the recorded and the auralized musical fragments, they proved that it is possible to develop reliable auralizations from virtual models [17].

Several studies have also focused on this coupling between subjective and objective parameters in the last decade or so. In 2012, researchers from the Polytechnic University of Valencia and the University of Valencia conducted a study to find a common vocabulary between acousticians and melomaniacs through a series of surveys. The results showed a common vocabulary between the two groups that can vary a lot depending on the interviewees’ training level [18]. That same year, another group of researchers from the same university addressed this issue by interviewing experts and non-experts. The study evaluated seventeen music venues using the Kansei Engineering methodology [19].

To this day, researchers continue working on parametrizing the subjective listening experience. This is the case of a study published in 2020 by researchers from the School of Architecture of Tianjin University. They evaluated listeners’ preferences in different Chinese performance buildings. The study showed differences in the preferences for different parameters depending on the evaluation system. The Chinese evaluation system places a greater emphasis on loudness and brightness when compared to the Western evaluation system [20].

Also, during the latter part of the 20th century, the acoustic of buildings started to gain attention in the heritage preservation of buildings. While architectural and historical features have long been recognized as essential elements of cultural heritage, the specific focus on the acoustic characteristics of buildings is relatively recent. One of the factors that led to this change was an increase in the number of restorations of historic theaters and concert halls. As awareness grew about the unique acoustic qualities of these spaces and their contribution to the overall cultural experience, efforts were made to preserve and restore them. Notable examples include the restoration of renowned venues like La Scala in Milan or the Concertgebouw in Amsterdam.

The recognition of the acoustics of buildings as an integral part of cultural heritage continues to evolve. Today, the preservation of soundscapes, the acoustic design of spaces, and the consideration of a building’s auditory history are increasingly valued in the broader conversation about cultural heritage conservation. While specific initiatives and guidelines may vary by region and organization, the acknowledgment of acoustic elements as cultural heritage reflects a growing appreciation for the multi-sensory aspects of historical places [21]. This has been an important topic for conversations about the restoration of Notre Dame, where not only do the visual aspects need to be restored but also the acoustics [9,22,23].

This study builds on previous works to analyze the acoustics of the renowned Banda Primitiva theater using subjective and objective parameters, surveys of musicians, and auralizations. The objective is to understand how the theater’s acoustics have evolved over the years and to unveil the reasons behind the preference of the regular audience for specific seats or areas of the venue, and if these tendencies are influenced by tradition rather than the current auditory experience. Moreover, it seeks to highlight the importance of architectural acoustics in Llíria and València’s cultural landscape and to find connections between the vocabulary used by musicians and acousticians in the local and global contexts. To the authors’ knowledge, there has not been any paper that analyzes the acoustic characteristics and the intangible cultural heritage of this important landmark venue in Valencian folk culture. Also, this study introduces the novelty of bridging the gap between the vocabulary used by musicians and acousticians in subjective acoustic evaluations.

The structure of this article follows the IMRAD format: Introduction, Methods, Results, and Discussion, as well as a conclusion section. The methodology section first describes the venue and then describes the methodology used for the objective and subjective evaluation in separate sections. In this case, the results and discussion sections are merged. This section is also divided into the objective and the subjective results. The objective results include the measurement results and the ray-tracing simulation process. The subjective results include the test results and the statistical evaluation of the tests conducted.

2. Materials and Methods

The methodology of this paper is composed of three parts. First, an architectonic and constructive analysis of the theater was necessary for the rest of the sections. Secondly, we carried out a study of the past and present objective acoustic parameters of the theater. Lastly, a subjective evaluation using auralizations was obtained from questionnaires.

2.1. Case Study

The theater under study is located inside the Banda Primitiva de Llíria facilities. Figure 1 depicts a section on the ground floor of the venue. It was built in 1951, based on architect Joaquín Rieta’s project design. The, at the time, innovative structural framework enabled a broad interior spatial arrangement comprising stalls and two balconies. After four decades of musical events, the theater underwent renovation in 1991. Its original design was respected, but the room’s seats and materials were changed [6]. In the first intervention, the seats, the upholstery, the flooring, and the wooden moldings of the plinth were renewed; in the second, the stage floor was changed, a platform lift was installed, new recording and lighting equipment was installed, as well as new curtains, and the dressing rooms were renovated; in the last one, the moldings of the foyer were gilded again, the flooring of the side areas was replaced, and new lamps were installed. The works were carried out by the architect José María Nevado and the quantity surveyor José Vicente Bori [8].

Currently, the venue has a seating capacity of 967 and features a staging area of 221.6 m², which includes the apron. The stage measures 11 m in width and 9.70 m from the stage opening, with an additional 3.80 m in the depth of the apron. The stage opening is 10.10 m wide per 6.60 m in height. The fly tower is 16.75 m high. The seating area consists of stalls covering a surface area of 468 m², along with two balconies; the first balcony spans 56.65 m², and the second spans 187.65 m².

Regarding the materials, most vertical walls and ceilings are rendered in plaster; they include large, padded surfaces covered in fabric and a wooden baseboard around the perimeter. The ceiling is made of plaster, both in the general room and under the balconies’ slabs. Plaster ceilings are the most common in theaters built in Spain during the 20th century [24,25]. A large chandelier is suspended from the ceiling in the center of the room. Several elements in the room are decorated with plaster, such as the first balcony’s baseboard, the stage opening’s arch, and the outer circle adjacent to the ceiling lamp. The lateral walls and the ceiling meet with a curved surface in the stall area. The baseboard of the balconies is made of plaster, with a wooden covering facing the stage. Some side walls and the back of the room are also made of plaster with an air cavity.

The floor of the stall area is made of wooden parquet, while the corridors are covered with carpeting. The first balcony floor is a two-level wooden platform in the area with independent seats near the stage and linoleum in the balcony. The stairs and the baseboard at the height of the access handrail are marble. The balcony has a painted mortar floor with a terrazzo tiled floor on the access staircase.

The first balcony has a mix of upholstered and medium-upholstered seats, with the upholstered ones located in the stalls and back seats and the medium-upholstered ones in the independent seats at the front. The balcony has medium-upholstered seats. Originally, the seats were handmade of wood by the members of the musical association. The replacement of the wooden chairs with the theater’s more upholstered seats today, the wood cladding on the corridors, and the fabric hanging are the most significant changes the theater has seen over the years.

The person with the cabin communicates with those in the room’s space through a small window, which remained closed during the measurement. The flooring is made of wood in the stage area, and the walls are covered with curtains. The stage opening is covered with a pleated velvet curtain.

2.2. Objective Evaluation

2.2.1. On-Site Measurements

The objective parameters of the venue were assessed by conducting a measurement following the guidelines of ISO 3382-1 (Figure 2) [26]. The impulse response was recorded in 20 positions in the audience plane, which is non-symmetrical in relation to the room axis. The sound source was placed in the middle of the apron (Figure 1). The sound source was located near the edge of the scene because, according to local musicians, in this specific venue, performers tend to play near the edge of the scenario, as they claim they hear themselves better. The intention was to replicate the actual performance conditions in the measurements. The microphone was placed in locations where there are people seated in attendance. In Figure 1, positions 11 and 13 are located in an area without permanent padded seats, but with regular movable velvet chairs. The measurements were taken with no audience. The process was carried out using DIRAC software v3.1, which uses sine sweeps as excitation signals to obtain the impulse response at each position. The software averages arithmetically the results of two consecutive measurements following the ISO 3382-1 guidelines [26]. A Brüel & Kjäer 4296 OmniPower omnidirectional speaker was used as the sound source, and a Shure KSM44A omnidirectional microphone was used to register the signal (Shure, Chicago, IL, USA). The sound source was placed 150 cm from the floor, and the microphone was 144 cm high. A single sound source position was used because if the materials and geometry of the room ensure an even distribution of sound absorption and diffusion properties, a single source position, combined with multiple microphone positions, can still provide a representative average of the acoustic parameters. The parameters evaluated from the impulse response were the EDT (early decay time), TR10 (reverberation time measured with a 10 dB decay), TR20 (reverberation time measured with a 20 dB decay), TR30 (reverberation time measured with a 30 dB decay), C80 (clarity; early sound energy that arrives at the listener within 80 ms), D50 (definition; the ratio of the early received sound energy, 0–50 ms after direct sound arrival, to the total energy), and STI (speech transmission index) [27].

2.2.2. Virtual Model of the Original Theater and Ray-Tracing Simulation

In this section of the methodology, we aimed to create a virtual model of the theater that accurately represented its original state. The theater was initially modeled in 3D using AutoCAD 2024 and subsequently imported into the ray-tracing simulation software ODEON to achieve this. The virtual model was then fine-tuned to replicate the acoustic parameters obtained with the on-site measurements. Based on the real theater, a simplified surface model was generated and calibrated by an iterative process based on T30 in an empty room. The audience was simulated by taking average absorption values of similar seats from recognized databases included in ODEON v18 software. Next, the model was adjusted to match the original 1957 design. The virtual model allows the impulse response of the theater to be obtained at different positions, which can be used to derive the theater’s acoustic parameters and auralizations. Since ODEON is a ray-tracing energy software, the results are more reliable at medium and high frequencies. This is compensated for by the tendency of the symphonic band music to have more high frequencies than low frequencies. The multi-criteria method of Higini Arau’s Hall of Merit Factor for Concert Audition was applied as an additional parameter. The result of this method is a single number ranging from 0 to 1 to rate the overall quality of the venue [28].

2.3. Subjective Evaluation

One hundred and three musicians were asked to complete a survey to compare the subjective acoustic parameters of the venue.

2.3.1. Recording and Auralization of Musical Samples

The survey was conducted using auralizations obtained by convoluting anechoic music with impulse responses obtained from ray-tracing virtual models. The use of auralizations was chosen as the only means to replicate, to a certain extent, the theater’s sound at the time of its construction. The auralizations produced from ray-tracing simulations are regarded as accurate as long as some considerations are taken into account. Reliability is subject to the limitations of computational capacity, the handling of reflections (both specular and diffuse), and the validation against real-world measurements. These factors introduce some level of approximation that users need to consider when interpreting the results from such simulations. Those aspects were resolved by conducting on-site measurements (Section 2.2.1) and conducting the simulation with the commercial software ODEON, extensively tested over the years by acousticians [29,30].

Auralizations obtained from ray-tracing models lack an accurate representation of the lower frequencies of the sound spectrum [31]. The musical fragments chosen were extracted from classical symphonic pieces to overcome this limitation. Symphonic music has a higher presence in mid to high parts of the spectrum. Besides that, this kind of music was also chosen because it is the most commonly performed in this venue. Other than the mentioned limitations of ray-tracing simulation software, the impulse responses completely characterize the behavior of rooms as time-invariant linear systems [29].

The musical pieces are “Pictures at an Exhibition” by Modest Petrovich Mussorgsky, and “Andante Cantabile” by Pyotr Ilyich Tchaikovsky. The pieces were recorded in the anechoic chamber in the School of Architecture of the Polytechnic University of Valencia. The recording was made with a single Shure KSM44A bidirectional microphone. There are many different ways proposed in the scientific literature to record sound to use in auralizations. There are no specific requirements for the number of speakers needed to accurately record it. In the case of this study, a stereo microphone was used to represent the binaural human listening experience. The literature shows that a stereo microphone can be used effectively in an anechoic recording environment to capture specific directional sound information [29].

The musical ensemble invited to participate was Vent-a-Cinc, composed of Josefina Martínez-Cano, flutist, Aitor Llimerá-Galduf, oboist, Santiago Pérez-Fernández, clarinetist, Miguel Antonio Puchol-Peñarrocha, bassoonist, and Vicent Navarro-Gimeno, a horn player. The French horn player Miguel Martínez-Megías also participated in the recordings.

Once the pieces were recorded, a 30 s sample of each was selected for testing. Short samples were chosen according to studies that point out that auditory memory is short [32,33]. These pieces were then convoluted with four different impulse responses: the original states of the stalls and the second balcony, and the current states in the same two positions. That made for a total of twelve fragments (

Table 1. Naming of the fragments in the PCT.

Melody	Location 1	Time	Fragment
1	GF	OS	1
1	GF	CE	2
1	FF	OS	3
1	FF	CE	4
2	GF	OS	5
2	GF	CE	6
2	FF	OS	7
2	FF	CE	8
3	GF	OS	9
3	GF	CE	10
3	FF	OS	11
3	FF	CE	12

¹ GF: stalls, FF: first floor, OS: original state, CE: current state.

). Those positions were chosen in response to the debate about the sound quality in the stalls and the second balcony, explained in the introduction.

2.3.2. Design of the Survey and Testing Process

As explained in previous sections, the subjective parameters of the theater were evaluated by conducting a survey that included 103 musicians. The development process of the survey can be found in previous work [34]. The methodology for subjective acoustic evaluation through surveys is well-established in the field of psychoacoustics and has been used in various studies to assess concert hall acoustics [27,35].

The questionnaire consisted of two parts. The first part gathered general information about the participants, such as their gender, age, musical background, hearing level, and the instrument they played (Figure 3). Collecting this demographic information helps control for variables that could affect subjective judgments, as recommended by previous research [36]. The second part was the subjective test, which was a Pair Comparison Test (PCT) (Figure 4) [37]. The PCT is a widely recognized method for subjective evaluation in acoustics due to its ability to minimize bias and facilitate direct comparisons between different acoustic environments [38]. The terminology for the testing was developed with the help of three professional musicians, Mª Teresa Barona-Royo (principal piccolo, Orchestra of València), José Miguel Martínez-Falomir (concertmaster, Municipal Symphonic Band of València), and Víctor Portolés-Alamà (viola, Orchestra of València). This collaboration ensured that the descriptors used were meaningful and relevant to the musicians, which is crucial for obtaining reliable subjective data [39].

All subjects participated in this study with the same equipment and under the same room conditions to reduce the influence of extraneous variables, such as equipment variability and environmental differences, on their judgments. To further reduce the influence of emotional factors, preliminary instructions were provided to the participants, emphasizing the focus on the acoustic attributes rather than personal preferences [40].

To mitigate hearing fatigue, the length of the survey was adjusted by conducting preliminary testing with university personnel. This step aligns with best practices in psychoacoustic testing, where test duration and breaks are carefully managed to maintain participant attention and accuracy [41].

The Pair Comparison Test (PCT) involved comparing twelve musical fragments consisting of three melodies and four room situations, resulting in twelve comparisons. This method allows for a comprehensive evaluation of different acoustic settings through a manageable number of comparisons, ensuring a balance between thoroughness and participant endurance [42]. Several attributes were evaluated during each comparison, as detailed in Figure 4. To ensure clear communication between musicians and acousticians, several terms that refer to the same attribute were used to clarify the specific concept. This approach follows the guidelines for developing perceptually meaningful attributes in subjective testing [43].

The consistency of the survey was tested with a Cronbach alpha analysis [37]. A principal component analysis (PCA) was conducted on the results of the subjective questionnaire [44].

3. Results and Discussion

3.1. Analysis of the Objective Evaluation

The objective parameters obtained after processing the on-site measurements can be seen in

Table 2. Objective parameters measured in situ in position 2 (stalls).

Parameters	Octave Bands (Hz)
	125	250	500	1k	2k	4k
EDT	1.14	1.11	1.07	1.13	0.83	0.81
TR30	1.43	1.45	1.21	1.21	1.17	1.00
C80	2.73	2.8	4.18	5.18	6.05	6.27
D50	0.33	0.23	0.46	0.63	0.69	0.6
STI	0.61

and

Table 3. Objective parameters measured in situ in position 19 (second balcony).

Parameters	Octave Bands (Hz)
	125	250	500	1k	2k	4k
EDT	1.12	0.97	0.85	1.09	0.88	0.71
TR30	1.53	1.38	1.20	1.22	1.18	1.01
C80	0.37	1.02	4.87	4.47	4.02	5.87
D50	0.4	0.46	0.64	0.58	0.57	0.64
STI	0.63

. Table 2 indicates the results of one of the measurement positions in the stalls, and Table 3 the results of one of the positions in the second balcony (position 19). These results are the ones used in the following sections for the subjective comparison. The full results of the measurements are available as Supplementary Materials.

3.1.1. Three-Dimensional ODEON Model Adjustment

The response surface method (RSM) was employed to adjust the virtual models of the theater to match real-life measurements of reverberation time (TR) [45]. As briefly introduced in Section 2.2.2, initially, 3D models were created in AutoCAD and imported into ODEON software for acoustic simulations. The absorption coefficients for various surfaces were sourced from the literature, lab tests, or the ODEON material database. Often, initial model calculations differed from actual measurements, necessitating adjustments.

The RSM was introduced to systematize this adjustment process, particularly for large-volume spaces. It allows simultaneous adjustments of two unknown surfaces by varying their absorption spectra and observing the effect on TR. The method involves defining a study region for each unknown surface by using initial bibliographic values (B1i, B2i) and applying increments (R1i, R2i) to form combinations of absorption coefficients.

For each frequency band, absorption coefficients of surfaces (X1i, X2i) are combined to create nine different absorption spectra. These combinations are then simulated in ODEON to calculate the resulting TR30 values. The results are compared to target TR values measured in situ. By plotting response surfaces and examining how variations in absorption spectra impact TR, the RSM helps identify which surface absorptions significantly affect the acoustic behavior, thus refining the virtual model to closely match the real-world acoustics. An extensive explanation of the methodology can be found in the doctoral thesis written by Blanca Pérez Aguilar [46]. The absorption coefficients applied to each surface can be seen in

Table 4. Absorption coefficients applied to each surface.

Zone	Materials	125 Hz	250 Hz	500 Hz	1 KHz	2 KHz	4 KHz
Around stage mouth, moldings and ceiling curves, parapet P1 zone 1a	Decorated plaster molding	0.13	0.13	0.25	0.28	0.30	0.30
Stage ceiling room background, parapet P2	Plaster on brick hollow sound	0.16	0.10	0.06	0.04	0.04	0.04
Stage and amphitheater flooring P1 zone 1a	Tongue-and-groove wooden flooring on reinforced concrete beams (wooden platform with great air depth)	0.40	0.30	0.20	0.17	0.15	0.10
PB flooring	Wooden flooring in a chamber (parquet on battens)	0.05	0.03	0.06	0.09	0.10	0.20
PB carpet	Carpet on wooden flooring	0.11	0.14	0.37	0.43	0.27	0.25
Amphitheater flooring P1, zone 2a	Linoleum or vinyl glued to concrete	0.02	0.02	0.03	0.04	0.04	0.05
Amphitheater flooring P2	Enamel-painted concrete	0.01	0.01	0.01	0.02	0.02	0.02
Floor stairs access P1 + baseboard and P2	Marble or polished tile	0.01	0.01	0.01	0.01	0.02	0.02
Vertical walls (hollow sound)	Plastering on plasterboard with an air chamber	0.29	0.10	0.05	0.04	0.07	0.09
Vertical walls (solid sound)	Plaster on brick	0.01	0.02	0.02	0.03	0.04	0.05
P1 parapet zone 2a, ground floor plinth, stage front	Thin wood (5–10 mm) forming an air chamber at the back	0.42	0.21	0.06	0.05	0.05	0.04
Door access to plants	Solid wood, 5 cm thick	0.01	0.05	0.05	0.04	0.04	0.04
Amphitheater seats 1 and 2	Medium upholstered armchairs	0.56	0.64	0.70	0.72	0.68	0.62
Silver seats	Heavily upholstered chairs	0.72	0.79	0.83	0.84	0.83	0.79
General roof and low overhang first amphitheater	13 mm plaster on 25 mm studs (with mineral wool)	0.28	0.20	0.12	0.07	0.05	0.07
Quilted fabric frames on walls	Quilted fabric frames on walls	0.17	0.50	0.57	0.55	0.55	0.50
Scene mouth	Scene mouth	0.15	0.18	0.37	0.45	0.50	0.57

. As shown in the table, the P1 parapet, the stage and amphitheater flooring, and the silver seats exhibit high absorption coefficients at 125 Hz. This high absorption is due to the construction of these surfaces, which are made of wood with an air chamber behind them. In acoustics, this configuration is known as a membrane, which typically has high absorption coefficients at low frequencies.

We modeled the theater in the original and the current states using ODEON, and the obtained results can be seen in Figure 4. The adjusted model does not differ more than 5% from the original on-site measurements, which is considered adequate in terms of the Just Noticeable Difference (JND) [31].

As shown in Figure 5, the venue in its original state had a longer reverberation time than nowadays. This effect is reduced when the seats are fully or partially occupied by the audience. In this case, there are no remarkable differences between the room with occupied seats in the past and the room in its current state. Possibly, the most significant difference from before refurbishment to nowadays would be noticed by musicians during the rehearsals with no audience. The C80 results can be seen in Figure 6.

3.1.2. Objective Evaluation of the Quality of the Venue at Two Instants in Its History

The multi-criteria method of Higini Arau’s Hall of Merit Factor for Concert Audition was applied to compare the quality levels of the current and the original halls [28]. The method evaluates the acoustic quality on a scale from 0 to 1 based on the acoustic variables considered. The variables considered are the medium reverberation time (T30mid), medium early decay time (EDTmid), clarity (C80), definition (D50), bass ratio (BR), and brilliance (Br). The mean values for each of the three hearing zones are presented. The rating obtained is above 0.8 for concert auditions. The original configuration scored slightly higher for music listening, especially in the stalls and amphitheater. This was due to the lack of absorption on the sides. The average results are presented in

Table 5. Average and standard deviation for Arau’s Merit Factor and the two instants for the concert hall.

	Current	Original
Stalls	0.852 (0.03)	0.861 (0.04)
Balcony	0.840 (0.09)	0.877 (0.07)
Second balcony	0.894 (0.03)	0.927 (0.01)

.

The full measurements taken in the venue and the simulation results are presented in as Supplementary Materials. The results show that the overall deviation in T30 of 0.07 s, which indicates that the acoustics in the theater are fairly homogeneous.

3.2. Analysis of the Subjective Evaluation

The test was carried out on one hundred and three subjects; a further four were considered null because they were unfinished. The percentages of women and men in the sample were almost equal. The age groups were reasonably distributed. More than half of the participants were professionals. All of the participants were either trained in music or music enthusiasts. Most of them considered themselves to have a good hearing capacity. Just over half of the participants played wind instruments. All participants younger than nineteen were music students at the conservatory. The main characteristics of the survey sample are summarized in

Table 6. Main characteristics of the survey sample.

Percentual Result	%
Gender:
Female	44.66
Male	54.37
Age:
11 to 19	17.48
20 to 29	16.50
30 to 39	16.50
40 to 49	19.42
50 to 59	21.36
60 to 69	7.77
Relationship with music and/or concert halls:
	Men	Women	Total	%
Professional	39	22	61	59.22
Student	9	16	25	24.27
Music Lover	8	8	16	15.53
Self-skill-level judgement:
Good	51	34	85	82.52
Average	5	10	15	14.56
Low	0	2	2	1.94
Type of instrument:
Wind	55.6
String	15.2
Percussion	23.2
Not specified	6.06

.

3.2.1. Consistency of the PCT

A Cronbach alpha analysis (internal consistency) of the compiled data proved the consistency of the PCT. The alpha was 0’886, which showed the internal consistency was good (almost excellent).

3.2.2. Results of Subjective Evaluations

In the subjective evaluation test, participants made three different comparisons, each repeated with three different melodies.

Table 7. Comparisons related to the fragments, questions, and melodies.

Comparison	Fragments		Question	Melody
Original vs. nowadays in the stalls	1	2	1	1
	5	6	3	2
	9	10	5	3
Original vs. nowadays in the second balcony	3	4	2	1
	7	8	4	2
	11	12	6	3
Stalls vs. second balcony nowadays	2	4	7	1
	6	8	8	2
	10	12	9	3

provides a detailed breakdown of these comparisons, including the corresponding questions, music fragments, and melodies.

A PCA was conducted for a comparison (Figure 7, Figure 8, Figure 9, Figure 10, Figure 11, Figure 12, Figure 13 and Figure 14). Figure 7 is an influence graph for comparison 1, comparing the original and current states of the stalls. It graphically represents the coefficients of each variable for the first component versus the coefficients for the second component. C1 corresponds with questions in which the melody is number 1, C3 corresponds with questions with melody 2, and C5 with questions with melody 3.

Moreover, PC1, corresponds to a preference for the original state. In contrast, the second component, PC2, reflects a preference for the current state. To illustrate this, consider Participant 77, located at point 0.0 on the two-dimensional graph. This participant consistently rates the heard fragments as having a similar acoustic quality.

In summary, PC1 and PC2 represent distinct factors influencing participants’ perceptions of the theater’s acoustic environment. Analyzing participants’ responses concerning these factors provides a better understanding of how they perceive the theater’s sound quality and how it relates to their preferences for either the original or the current state. A1-6 and G are the relevant parameters investigated in the questionnaire.

The graph shows the degree of influence of each variable on the two components. The arrows represent the direction and magnitude of the influence, and the numbers correspond to the participants. The closer a variable is to 1 on a component, the stronger its influence. For example, Question 3 has a relatively weak influence on component 2, with a maximum value of 0.2. Overall, the graph provides insights into how each variable contributes to the participant’s perception of the theater’s acoustic environment.

Figure 8 refers to comparison 1. In this case, it involves analyzing the type of instrument and its effect on the preference for the current state or the original state in the theater stalls. The percentages of the PCA are included, which show that components 1 and 2 explain 33.6% of the variance. However, it is important to note that the fragments analyzed are very similar due to the similarity between the two rooms. Despite this, most participants fall within the range of 0.0–0.1 in Component 1 and between −0.1 and +0.1 in Component 2, indicating that they do not have a clear preference between the states. A small group is located at the top left of the graph, showing a clear preference for Component 2 over Component 1.

Figure 9 also corresponds to comparison 1. In this case, the participants have been classified according to their level of music involvement, that is, whether they are professionals, enthusiasts, or students. The results show that Components 1 and 2 explain the result of 33% variance. Also, most of the responses orbit around 0.0.

Figure 10, Figure 11 and Figure 12 refer to comparison 2, which analyzes the differences in perception between the current state and the original state of the theater in the second balcony. Figure 9 is an influence graph. C2 refers to melody 1, C4 to melody 2, and C6 to melody 3. Component 1, PC1, refers to the preference for the original state, and Component 2, PC2, refers to the preference for the current state of the theater. These questions have a certain influence over Component 1. Figure 11 and Figure 12 are equivalent to Figure 8 and Figure 9 but refer to comparison 2 instead of 1. Components 1 and 2 explain 35.6% of the variance in this case. In both Figure 11 and Figure 12, most points revolve around zero, which indicates there is no clear preference for either the original or the current state of the theater in the second balcony. That suggests that neither the type of instrument played nor the level of training plays a significant role in this comparison.

Figure 13, Figure 14 and Figure 15 are based on comparison 3, which compares the stalls and the second balcony in the current state. In Figure 12, 789 represents comparison 3, C7 melody 1, C8 melody 2, and C9 melody 3. In this case, PC1 refers to the preference for the stalls, and PC2 refers to the preference for the second balcony. These questions have a certain negative influence over Component 1. Figure 13 shows the distribution of preference according to the type of instrument and Figure 15 according to the training level of the interviewees. Although the results still revolve around zero in both figures, there is a certain dispersion towards PC2, which indicates a subtle preference for the second balcony.

In summary, the PCA tests show that the main components do not determine a significant percentage of the variance and that the interviewees showed a certain level of dispersion in their responses. However, as can be seen in Figure 6, Figure 9 and Figure 12, we can interpret the results as meaningful for the following reasons. Firstly, in terms of their clustering, the majority of the samples are clustered around the center, indicating similar characteristics, with some samples consistently appearing as outliers. Secondly, in terms of the variable influence, variables have long arrows pointing in similar directions across different PCA plots, indicating they are significant contributors to the observed variance.

The comparisons below are expressed in percentages (

Table 8. Comparison 1, original vs. current state in the stalls.

	%
	A1	A2	A3	A4	A5	A6	G
Question 1
COMP_1_CRIT_1	63.6	37.4	48.5	36.4	17.2	26.3	22.2
COMP_1_CRIT_2	25.3	54.5	24.2	49.5	65.7	53.5	62.6
COMP_1_CRIT_0	11.1	8.08	27.3	14.1	17.2	20.2	15.2
Question 3
COMP_2_CRIT_1	64.6	20.2	17.2	62.6	29.3	19.2	27.3
COMP_2_CRIT_2	15.2	66.7	64.6	17.2	41.4	49.5	50.5
COMP_2_CRIT_0	20.2	13.1	18.2	20.2	29.3	31.3	22.2
Question 5
COMP_3_CRIT_1	38.4	39.4	35.4	30.3	22.2	35.4	23.2
COMP_3_CRIT_2	42.4	51.5	39.4	49.5	58.6	47.5	60.6
COMP_3_CRIT_0	19.2	9.09	25.3	20.2	19.2	17.2	16.2
Average per criteria
COMP_135_CRIT_1	55.6	32.3	33.7	43.1	22.9	26.9	24.2
COMP_135_CRIT_2	27.6	57.6	42.8	38.7	55.2	50.2	57.9
COMP_135_CRIT_0	16.8	10.1	23.6	18.2	21.9	22.9	17.8

CRIT_1: prefers first fragment of the comparison, CRIT_2: prefers second fragment of the comparison, CRIT_0: does not know or seems the same. A1: direct sound, A2: sweet, round, warmth, A3: intimacy, A4: brilliance, A5: reverberance, A6: ensemble, G: preferred fragment.

). When comparing the theater of origin to the theater nowadays in the stalls, the survey respondents prefer, in general, the sound nowadays, with a 57.9% average. The participants perceive the direct sound (A1) and, in a lower proportion, brilliance (A4) to be more intense in the past. At the same time, warmth (A2), intimacy (A3), reverberance (A5), and ensemble (A6) are sensed to be more present nowadays. In the second comparison (

Table 9. Comparison 2, original and current states in the second balcony.

	A1	A2	A3	A4	A5	A6	G
Question 2
COMP_1_CRIT_1	46.5	39.4	44.4	32.3	10.1	33.3	28.3
COMP_1_CRIT_2	35.4	46.5	20.2	52.5	72.7	45.5	50.5
COMP_1_CRIT_0	18.2	14.1	35.4	15.2	17.2	21.2	21.2
Question 4
COMP_2_CRIT_1	47.5	29.3	38.4	36.4	20.2	21.2	21.2
COMP_2_CRIT_2	30.3	53.5	36.4	51.5	60.6	58.6	62.6
COMP_2_CRIT_0	22.2	17.2	25.3	12.1	19.2	20.2	16.2
Question 6
COMP_3_CRIT_1	30.3	49.5	64.6	22.2	12.1	38.4	32.3
COMP_3_CRIT_2	56.6	44.4	20.2	71.7	75.8	43.4	57.6
COMP_3_CRIT_0	13.1	6.06	15.2	6.06	12.1	18.2	10.1
Average per criteria
COMP_246_CRIT_1	41.4	39.4	49.2	30.3	14.1	31	27.3
COMP_246_CRIT_2	40.7	48.1	25.6	58.6	69.7	49.2	56.9
COMP_246_CRIT_0	17.8	12.5	25.3	11.1	16.2	19.9	15.8

CRIT_1: prefers first fragment of the comparison, CRIT_2: prefers second fragment of the comparison, CRIT_0: does not know or seems the same. A1: direct sound, A2: sweet, round, warmth, A3: intimacy, A4: brilliance, A5: reverberance, A6: ensemble, G: preferred fragment.

), of the original building versus the theater nowadays in the second balcony, 56.9% of the survey contestants prefer, in general, the fragments that correspond to the theater nowadays. The direct sound is perceived to be more intense in the original state in the first two questions but the opposite in the third (

Table 10. Comparison 3, current state in in the stalls vs. the second balcony.

	A1	A2	A3	A4	A5	A6	G
Question 7
COMP_1_CRIT_1	31.31	57.58	53.54	16.16	12.12	38.38	34.34
COMP_1_CRIT_2	52.53	27.27	16.16	64.65	64.65	34.34	41.41
COMP_1_CRIT_0	16.16	15.15	30.30	19.19	23.23	27.27	24.24
Question 8
COMP_2_CRIT_1	31.31	40.40	50.51	21.21	13.13	32.32	26.26
COMP_2_CRIT_2	46.46	42.42	23.23	58.59	63.64	43.43	48.48
COMP_2_CRIT_0	22.22	17.17	26.26	20.20	23.23	24.24	25.25
Question 9
COMP_3_CRIT_1	39.39	59.60	60.61	27.27	22.22	48.48	46.46
COMP_3_CRIT_2	48.48	36.36	22.22	62.63	64.65	36.36	41.41
COMP_3_CRIT_0	12.12	4.04	17.17	10.10	13.13	15.15	12.12
Mean per criteria
COMP_789_CRIT_1	34	52.5	54.9	21.5	15.8	39.7	35.7
COMP_789_CRIT_2	49.2	35.4	20.5	62	64.3	38	43.8
COMP_789_CRIT_0	16.8	12.1	24.6	16.5	19.9	22.2	20.5

CRIT_1: prefers first fragment of the comparison, CRIT_2: prefers second fragment of the comparison, CRIT_0: does not know or seems the same. A1: direct sound, A2: sweet, round, warmth, A3: intimacy, A4: brilliance, A5: reverberance, A6: ensemble, G: preferred fragment.

), resulting in a draw. The attributes warmth (A2), brilliance (A4), reverberance (A5), and ensemble (A6) are perceived to be more intense in the theater nowadays. In comparison, intimacy (A3) is scored more for the theater in the past.

When listening to two music fragments that correspond to the theater nowadays in the stalls and the second balcony, 43.8% of the survey contestants preferred the fragment corresponding to the second balcony. In this comparison, warmth (A2) and intimacy (A3) are perceived to be more intense in the stalls, while direct sound (A1), brilliance (A4), and reverberance (A5) are sensed to have more presence in the second balcony. Almost the same percentage of participants found the attribute ensemble (A6) more intense in the stalls as those who did in the second balcony.

Figure 16 shows all the comparisons. The attribute reverberance (A5) is the one that reflects the preferences of the survey contestants more consistently. Inconsistencies in the preferences expressed in the other attributes could be due to confusion in the meaning of the attributes (what they are referring to), the difficulty of the hearing test itself (because the fragments were very similar), or simply personal preferences depending on what piece was being played.

In general, the obtained subjective results can be summarized as follows:

Comparison 1 (Original vs. Current State in the Stalls):

• PC1 (Preference for Original State): Majority clustered around 0.0–0.1.
• PC2 (Preference for Current State): Some dispersion towards PC2 indicating a subtle preference for the current state.

Comparison 2 (Original vs. Current State in the Second Balcony):

• Components 1 and 2 explain 35.6% of the variance.
• No clear preference for either state; most points revolve around zero.

Comparison 3 (Stalls vs. Second Balcony Nowadays):

• PC1 (Preference for Stalls) vs. PC2 (Preference for Second Balcony).
• Subtle preference for the second balcony indicated by dispersion towards PC2.

The arrows in the graphs show the direction and magnitude of each variable’s influence. Variables closer to 1 on a component axis have a stronger influence. For example, Question 3 has a relatively weak influence on PC2 with a maximum value of 0.2.

Detailed Analysis:

• Figure 7, Figure 8, Figure 9, Figure 10, Figure 11, Figure 12, Figure 13 and Figure 14 show further analyses, such as the impact of the type of instrument and the level of musical involvement on preferences.
• Figure 15 summarizes all comparisons. Attributes like reverberance (A5) consistently reflect preferences, while inconsistencies in other attributes might be due to confusion, the difficulty of the test, or personal preferences.

Preferences and Attributes:

• Original vs. Current State in the Stalls: 57.9% prefer the current state. Direct sound (A1) and brilliance (A4) were more intense in the past, while warmth (A2), intimacy (A3), reverberance (A5), and ensemble (A6) are more present nowadays.
• Original vs. Current State in the Second Balcony: 56.9% prefer the current state. Direct sound (A1) was more intense in the original state for the first two questions but not the third.
• Stalls vs. Second Balcony Nowadays: 43.8% prefer the second balcony. Warmth (A2) and intimacy (A3) were more intense in the stalls, while direct sound (A1), brilliance (A4), and reverberance (A5) had more presence in the second balcony.

4. Conclusions

In this research, the objective and subjective acoustic parameters of the theater of the Banda Primitiva de Llíria have been analyzed. The main idea of the work was to assess the differences with the original state of the theater at the time of its construction. This paper has presented the unique cultural and acoustic heritage of the Banda Primitiva de Llíria theater, a historically significant venue designed by Joaquín Rieta. This dual focus on cultural history and acoustic analysis sets the research apart. This study employed objective acoustic measurements and subjective evaluations through surveys and auralizations. Our comprehensive approach allows for a multifaceted understanding of the theater’s acoustic properties and audience preferences. The research included a virtual reconstruction of the theater’s original state, enabling a comparison between its past and current acoustic performance. This historical perspective was crucial for assessing the impact of renovations on the theater’s acoustics. By surveying 103 musicians and analyzing their responses using PCA, this study collected empirical data on how professional and amateur musicians perceive acoustic changes. This adds a layer of practical relevance and real-world applicability to the findings. Also, the originality of this paper stems from its hybrid methodology, combining auralizations in past and present environments, allowing for a comparison between indoor soundscapes.

After finishing this study, the following conclusions can be drawn:

• The original state of the theater does not differ significantly from the current state in terms of its objective acoustic parameters. However, there is a slight decrease in the overall reverberation time in the current state compared to the original state of the theater. This is primarily because the original chairs were made of wood. Currently, the theater has padded seats and fabric hangings.
• The objective tests show that the theater’s original and current configurations are adequate for the kind of performances that take place in the venue.
• The lower RT of the current state of the theater can be positive for theater representations and symphonic music performances. Also, the results are uniform across all the auditorium areas and little influenced by the occupation percentage.
• According to the merit factor score, the quality of the original state of the theater was slightly higher than that now. While the original theater scored up to 0.927, the current state reached 0.894. Any score higher than 0.8 is considered an excellent result.
• The theater can be compared, both acoustically and culturally, to theaters like the Vilamarta theater in de Jerez de la Frontera. Built in the year 1928, this theater has a comparable design, purpose, and objective acoustic parameters to the one studied in this article, showing they are both versatile and adequate for voice performances and symphonic band concerts [47,48].
• The subjective tests showed that 57.4% preferred the current state of the theater. Also, 43.8% selected the current state of the theater at the second balcony as their preferred configuration, which supports the widespread belief that the second balcony provides an overall better auditory experience.
• Despite the slight majority of the sample preferring the second balcony, the preference for it could not be explained by the PCA tests, which did not explain a significant percentage of the variance.
• That can also be extended to the differences between the original state of the theater and its current state.
• The main change in replacing the wooden seats with padded ones has been in reducing the differences in acoustic parameters depending on the attendance.
• The results indicate that the widespread belief that the second balcony has a much better acoustic condition than the rest of the theater cannot be supported by objective or subjective evidence.
• Although the subjective and objective parameters are difficult to compare, it can be said that they correlate well with each other due to the overall satisfaction of the listeners that conducted the test and the high results obtained in the merit factor.
• Overall, this paper provides a thorough and original analysis of the Banda Primitiva de Llíria theater, combining historical reconstruction, objective measurements, and subjective evaluations. The PCA results and survey data collectively suggest that while there is a slight preference for the current state of the theater, the differences are subtle and not statistically overwhelming. This nuanced understanding contributes to the field of architectural acoustics and cultural heritage preservation.

This study has some limitations that are worth mentioning. First, all the instruments played in the musical pieces used for the auralizations were wind instruments. Secondly, the differences between the theater’s original and current states are relatively small. That made it difficult even for well-trained musicians to distinguish the small nuances in each musical fragment. Interpretations of the acoustic qualities of the different positions in the theater were subjective, and some of the preferences of people who remember the former state of the theater may arise from sensory experiences unrelated to music and pure perception. Sensory preferences play a tremendous role in our perception of concert halls, as was discussed by Lokki [49].