【Abstract】The lecture explains the methods behind this new concept that combining spatial sound effects and room enhancementand ,and shows examples of temporary and fixed installations with the Stagetec Vivace system.
　　【Key Words】theater; sound system; spatial sound effects; room enhancement; Vivace system
　　Like the recent step of the cinemas into 3D video projection, the audio systems for theaters and event halls are evolving into a new dimension. The concept of a frontal sound reinforcement augmented by additional surround speakers is going to give way to more sophisticated solutions. The loudspeakers around the stage are used to create a natural localization of the sound sources while a hemisphere of loudspeakers enveloping the audience creates the desired acoustical impression of the sound. This means that direct sound and reverberance are no longer mixed together in one signal, which enhances naturalness as well as the bandwidth of possible expressions. In combination with the classical abilities of a room enhancement system, the new theater sound reaches unknown dimensions.
　　The lecture explains the methods behind this new concept and shows examples of temporary and fixed installations with the Stagetec Vivace system.
　　1 Introduction
　　One of the big challenges for theater sound and in general for all entertainment venues is to be appreciably better than the audio video systems the audience uses at home. For the cinema industry this issue is most vital, so they are going ahead in this permanent race. This means they have to use highest resolution for video projection and are going to implement additional loudspeakers in the ceiling with an object-based positioning of the sounds.
　　This lecture shall present some audio features which help entertainment venues to keep the home entertainment systems at bay. The most important points for this are creating a real acoustical room simulation and preserving the natural localization of all sound sources. This makes a huge difference in the perception and for home entertainment it is a hard way to follow.
　　2 Room simulation
　　The aim of the acoustical room simulation in theaters and event venues is to create the acoustical impression of a realistic environment for sound sources of all kinds. It is not like in conventional sound reinforcement situations where the sound with some additive reverberation is emitted by the frontal speakers and in best cases is enhanced by some additional surround signals. It is much more: The direct sound is reproduced without any additives. A net of loudspeakers evenly distributed around and above the audience is creating reflections and reverberation matching the direct sound and thus forming a realistic and believable acoustical environment. With this the audience is sitting in a real and naturalistic room and not in front of some reverberant music. 　　This has some interesting and helpful implications: The more natural the created sound is, the more involved and engaged is the audience. This step is normally happening unconsciously. The amount of reverberation the sound designer can add to the sound without blurring and spoiling the informational content of the original signal is much stronger than in a traditional setup. Music is suddenly developing freely through the hall like in a real concert hall. The sound can be formed exploiting all the unconscious functionality the human brain uses to get a meaning of the environment the individual is sitting in. This creates new artistic possibilities and much more flexibility for the sound design. And a drawback of standard stereo or surround setups suddenly disappears: There is no sweetspot anymore. The area where the intentions of the sound designer can be heard without corruption is only restricted by the coverage of the whole loudspeaker net.
　　How can this be done? The first step is the step from a channel-based sound to an object-based sound. The sound is not anymore saved in distinct channels for each loudspeaker but it is saved as the original input signal together with object information about the geometrical position in the room and the properties of the desired simulated acoustic situation.
　　The sound system needs to know the geometrical position of all loudspeakers and transforms the original signal into reflections of virtual walls and ceiling elements and parts of a diffuse reverberation. All together it forms a natural sound field.
　　The basis for creating these reflections and reverberation are measured impulse responses which are convolved with the original signal. These measurements contain all natural acoustic properties of the real hall they were performed in.
　　The first part of the impulse response is formed by the early reflections which carry the clue to properties like the geometrical proportions of the hall or the distance to the sound source. These reflections have to be rearranged on the timeline with respect to the positions of the loudspeakers. This means that when creating the signal for each individual loudspeaker the system has to configure the delays of the reflections in a way that all audience seats get the desired time structure of early reflections. Also the direction is important from which the reflected energy arrives at the listeners’ position. With this means you can broaden the sound source or support the energy of the source. 　　The late part of the impulse response can be handled more relaxed. In most cases the aim is the homogeneous sound field of a diffuse hall. The acoustics of all beloved concert halls end up in such a uniform and even distribution of decaying sound energy. Important for recreating this sound field is the uniform directional distribution and the diffuse conditions. To accomplish this, we need a net of loudspeakers with individual signals all around the audience. From each seat you should preferably see a hemisphere of loudspeakers which supply sound energy to your seat. To reduce the necessary amount of loudspeakers the used speakers should have a broad and uniform directivity.
　　As long as the distances from the loudspeakers to the audience are in a range which is comparable to the free paths of sound in the intended effect room, the sound design is absolutely flexible. If the real hall is much bigger than the simulated hall you want to create, it is more difficult. If it is possible to install a loudspeaker setup which divides the audience area into smaller sub-systems for which a uniform directional coverage can be achieved, the impression of a small hall can be achieved. Moreover, the visual impression sets a limit to what can be done. It is hard to create a naturally sounding illusion if the visual evidence is contradicting. If the minds of the listeners are focused on the show by an appropriate light design or video projections, the flexibility for the sound design increases again.
　　Another important restriction which has to be borne in mind are the natural acoustic conditions of the event location. Since you can only add sound energy, it is hard to create the illusion of a dry room in a really reverberant event hall. A sound-reflecting ceiling will probably ruin the sound design of an open-air scene with an open sky.
　　3 Localization
　　A really important point for the perceived quality of a sound reinforcement is a correct localization of the sound sources. When all actors on stage sit in the loudspeaker cabinet of the main PA when you close your eyes, there is a mismatch between optical and acoustical impression. Over the years we got used to many drawbacks but if you compare the results to a more sophisticated setup where the actors can be heard from where they can be seen, the advantages are evident. With this you can eliminate in the minds of the audience the existence of a sound reinforcement at all. And to be able to separate sound sources at different positions without an additional visual clue is a huge help to transport the contents of the event. The human brain is perfect in distinguishing sound sources coming from different directions, the so-called cocktail party effect. So let us exploit this ability! 　　The simplest way to achieve a sound reinforcement which preserves the correct localization is to use a loudspeaker at the position of the sound source. If the audience is too big to be covered by the directivity of one single loudspeaker, more loudspeakers for different parts of the audience may be used. In many cases the energy you can supply with these loudspeakers on stage is rather limited. An uncomfortable sound level in the first rows of the audience, feedback considerations and limitations by the stage scenery are reasons for this.
　　To get enough sound energy also to the rear parts of the audience, additional loudspeakers have to be used which cannot be located at the position of the sound source. To a certain limit it is possible to let the natural sound source itself or a loudspeaker nearby on stage define the position and add the necessary sound energy with other loudspeakers. This concept is commonly used to let the sound come from the stage level instead of the ceiling. But it can also be very helpful for creating a correct horizontal localization.
　　The main point is to ensure that the sound of the supporting loudspeakers arrives in a certain time window behind the direct sound that defines the audible position of the source. The tolerable time delay depends on the kind of the audio signal but for most purposes a range of 5 – 25 ms is a good value. The delicate limit is the starting value of the range. When the signal of the supporting loudspeaker reaches the listener nearly at the same time as the direct sound, the localization suddenly jumps to the loudspeaker.
　　For delay calculation not only the main coverage area of the loudspeakers has to be taken into account. Also in areas where the loudspeaker doesn’t deliver helpful high frequencies anymore the early arrival of this residual sound can shift or spoil the localization as the depicted example shows.
　　Figure 1: Delay for the main coverage area of the loudspeaker
　　Figure 2: Additional delay outside of the coverage area of the loudspeaker
　　There are two ways to get the input signal for the reinforcement. The first is to use fixed microphones which cover a certain area on stage. The appropriate delay between this microphone and each loudspeaker is a fixed value which is determined by the worst case position of the actor in the coverage area on stage and the listener in the extended coverage area of the loudspeaker.
　　The second possibility is to use wireless microphones. In this case, the delays for the loudspeakers have to be flexible and need to be adjusted when the position of the actor changes. The audio system needs to know the position of the sound source on stage. Depending on the kind of sound source, it might be possible to get this position by feedback information from some scenery elements or to save a predefined path for an actor. In most cases the position of the actor has to be changed or at least monitored by hand, for example on a touch screen. A very helpful means in this context is the usage of a tracking system which sends the position of the actor to the audio system. 　　4 Room enhancement
　　The third element in new theater sound is the usage of an electro-acoustical room enhancement system. Depending on the kind of system, the working principle can be similar to the described room simulation with its very high flexibility and naturalness. The significant difference to the room simulation is that the audio system has to cope with the feedback slope between the microphones and the loudspeakers. Since the microphones need to be at a distance of at least 3-4 m from the musicians to avoid the pick-up of unnatural near field sound and to get a well-balanced orchestra sound, there could be considerable feedback energy.
　　Depending on the type of system, this unwanted feedback slope is suppressed by a narrow directivity of the microphones or by a time variance which slowly and inaudibly changes the transfer function.
　　Some systems use the feedback slope between the loudspeakers and additional numerous microphones in the hall to create reverberation. During the sound decay the signal passes the microphone loudspeaker combination about 50-100 times. Even if additional early reflections are created electronically in some of these systems, the achievable flexibility in designing the desired acoustical conditions is much more limited with this approach.
　　5 Examples
　　The following two examples show the practical application of the described possibilities.
　　(1)Felsenreitschule of the Salzburg Festival
　　The Felsenreitschule in Salzburg, Austria with its stage rear wall carved out of a rock and its retractable stage roof is an absolutely unique venue among the major opera houses of international renown. In addition to opera and concert productions out of the classical repertoire, the venue also hosts big modern opera productions during the Salzburg festival.
　　Figure 3: Felsenreitschule, view to the stage
　　Figure 4: Felsenreitschule, audience area
　　The hall has a fixed installation of a Stagetec Vivace system which can be used for fine-tuning the sound perfectly to the respective audience, orchestra size and type of performance. It contains 72 fixed loudspeakers and approx. 16 variable loudspeakers which can be added in the scenery. The microphones can be freely positioned above the stage with the help of flexible telescopic bars which are installed on the lighting bridge.
　　Figure 5: Felsenreitschule, roof above the stage opened
　　Especially in the acclaimed modern opera productions of the Festival, there is a growing necessity of sound effects. These effects can be positioned as a sound object and moved around the audience area. Matching the artistic requirements, the effects can also be augmented by an effect room. This additional simulated room corresponds to the position of the sound object, thus augmenting the subjective impact of the effect. 　　(2)Oberammergau
　　In the little village of Oberammergau in Germany, there is a very special theatrical performance which shows the last days in the life of Jesus Christ. This passion play has been performed by inhabitants of the village every 10 years for about 350 years. Since 100 years, there exists a dedicated passion play theater with approx. 5,000 seats to host the 500,000 visitors in those years when the passion play is performed.
　　Figure 6: Passion play theater Oberammergau, view from the stage
　　Figure 7: Passion play theater Oberammergau, stage
　　A special challenge for this performance is the large stage which is populated by up to 800 actors. In the face of the long history of the event, the reinforcement of the non-professional actors should never be perceivable as electro acoustics. So it was absolutely necessary to realize a system which yields a perfect localization of the actors. On a broad stage with so many actors, it is of additional importance to hear who of them is speaking at the moment.
　　With the Vivace system a setup was realized which preserves the correct localization for all actors on stage via a set of fixed microphones and numerous loudspeakers for audience seats at a small, middle and large distance. As described above, the settings for each loudspeaker were determined for the coverage area of each microphone with the help of the Vivace simulation tools.
　　Figure 8: Tracking actors to get the correct localization when using wireless microphones
　　In addition, the room enhancement possibilities of the system were used to strengthen the reverberation level for the interlude of choir, orchestra and soloists. Due to the huge volume of the venue the reverberation time was more than long enough even with fully occupied audience areas. To make the reverberation and room response perceptible also during the music, a higher reverberation level with a slightly lower early reverberation time was applied.
　　In the years between the passion plays, the venue is also used for a popular production of a professional Bavarian theater. For this performance, the actors are equipped with wireless microphones and senders of a tracking system. The antennae of the tracking system detect the position of the actors via high-frequency pulses and transmit them to the Vivace system. The stage is divided into a grid of source areas. When the actor is standing inside one of these source areas his signal is fed into the loudspeakers with the appropriate delay, level and filtering. When the actor is moving to the next source area, a smooth transition between the respective settings of the two areas is realized.