Arnaud Laborie is the founder and CEO of Trinnov, one of the three most innovative French companies dedicated to audio. In his precise schedule, he very kindly gave us the time for this long and fascinating interview.
What is your background? Did you choose it according to your passion? Or did your training spark this passion?
Generally speaking, I am naturally attracted to scientific and technological fields and particularly those that touch on emotion and aesthetics, for example architecture, the automobile industry, design and the audiovisual industry. This led me to a classical training as an electronics and computer engineer (EFREI) completed by a Master in signal processing and acoustics (Télécom Paris) at IRCAM, the acoustics and music research laboratory founded by Pierre Boulez. The course came by passion without it being a really calculated choice. On the contrary, it can be said that the training brought revelations about the way things work which fed my passion for audio. At the EFREI I was in charge of the hifi club where the idea is to implement in a playful way the theoretical teachings through the conception and realization of different audio equipment: amplifiers, mixing tables, car radios, loudspeakers of all kinds (2-3-4 channels, closed, bass-reflex, amplified, compact, sound system…). These projects were moreover rather original. One of them, an innovative 4-way active linear phase system, had been presented to the AFDERS association and published with praise in the magazine HiFi Vidéo. In parallel to the emergence of the Trinnov project, I took a Master’s degree in Innovative Business Management (Paris IX Dauphine). Here it is the opposite, a choice totally calculated to favour Trinnov’s chances of success.
How was Trinnov born?
At the end of my engineering studies, I understood that I wanted to work for progress in the field of audio. I started from the fundamental question: what are the most promising opportunities to make a significant leap forward in sound quality. At the time it was the year 2000, with the launch of high-resolution audio: SACD and DVD-Audio. The idea was that the more precise the audio signal, the higher the sound quality. But once you get to 24 bits and 192 kHz, it is clear that any further progress in the same direction would bring insignificant hearing benefit. Conversely, the spatial aspect of sound was drastically underdeveloped. I presented this idea to 2 of my best friends, Sébastien Montoya and Rémy Bruno, both equally passionate about sound and music. Sébastien Montoya, a friend from EFREI promotion is an engineer in Electronics and Computer Science, and Rémy Bruno, met at IRCAM, is an engineer in Electronics and Signal Processing (Supélec). They immediately joined the project and together we decided to explore the new field of sound spatialization with the possibility of creating a company if the results are conclusive. We are at the end of 2000 and Trinnov has just been born. From the beginning, the company has been dedicated to 3D immersive sound and the acoustic optimisation of listening rooms. In both cases, it involves using digital processors and loudspeakers to control the way sound waves propagate in the space where the listener is. This is where our name 3D innovation or Trinnov comes from. In June 2003, the first technologies are developed, patents filed, scientific reputa-tion established with the first scientific publications (at the Audio Engineering Society, AES) and we obtain the support of the Ministry of Research. The company is officially created and we work on the development of the first products.
Why did you choose to develop this type of application in particular?
As I said, the primary motivation is to improve the quality of the sound experience. As soon as the quality of the signals was fully mastered thanks to the high resolution, the new field to be explored is the spatialization of sound, high spatial resolution. The importance of mastering the spatial dimension of sound is quite easy to understand. Just listen to the sound reproduced by a stereo or even 5.1 system and then compare it with natural listening, without an audio system, for example at an acoustic concert or simply in the street. In a natural listening situation, we are exposed to sounds coming from all directions, be it sound sources (musical instruments, cars, airplanes…) or the multiple secondary sources generated by the room acoustics. There are thousands (or even millions) of sound waves that reach us from all possible directions (left, right, front, back, top, bottom) and our auditory system is trained to analyse them in every detail: locate the direction of each source in 3D (azimuth and elevation), analyse the direction of each reflection of the room acoustics to deduce the distance of each source, the size of the room, the materials that make it up…. Wherever you are, simply close your eyes and try to mentally reconstruct your environment, find out where the sources are, at what distance, how the room is… you will be surprised by the quantity and accuracy of the information. That’s the spatial aspect of sound. If this information is not reproduced realistically enough by a reproduction system, the brain is lost and the reproduction is not credible, the experience is not «high-fidelity». From then on, we understand the new possibilities that open up if we manage to control sound waves from digital audio processors: the mastery of sound spatialization, the acoustic optimization of listening places (studios, living rooms, movie theaters), but also sound recording, post-production tools…
Had there been any attempts in this sector before you? How are they different from yours? In what way are you more innovative?
As far as sound wave control is concerned, there have indeed been some attempts in cutting-edge research laboratories. In the 1980s, a technology called Ambisonic was explored but although it was 3D (i.e. it could represent all directions in space) it was too limited in spatial resolution. In the 1990s, Wave Field Synthesis technology was able to achieve high spatial resolution but was limited to 2D (i.e. could not reproduce directions in elevation) and required a large number of speakers (about 100) to reconstruct wave fronts. The innovative side of Trinnov is twofold: on the one hand, to combine the 3D side and high spatial resolution, and on the other hand, to bring these advanced technologies to a commercial product. In terms of Acoustic Optimization of listening locations, there was an attempt by SigTech in the late 90’s to do more than just equalization by introducing a time domain correction in addition to the classical frequency domain correction. But Trinnov takes the concept further by bringing 3 innovations: 1/ that what is measured is the visible result of a more complex phenomenon corresponding to the propagation of sound waves in 3D in the listening room, 2/ that an acoustic diagnosis must be made to go back to the root cause of each acoustic problem and 3/ that each acoustic problem must be solved with a dedicated solution.
How does the Trinnov «diagnosis» work?
In terms of acoustic optimisation, Trinnov brings a fundamental innovation. It is not simply a matter of obtaining a desired response at a listening point by frequency equalization, i.e. by attenuating the frequencies that are over-amplified by the room and conversely by amplifying the attenuated frequencies. In reality, these dents and bumps can have many different origins: a loudspeaker listened to off-axis, a reflection against a wall, resonance, absorption or drag in the reverberation. Clearly these acoustic problems are different and do not produce the same effects when listening. They cannot be corrected in the same way. Therefore it is important to understand what is going on, i.e. first identify the cause of each problem before deciding how to correct it and sometimes even give up correcting it when this is not possible. This is exactly what Trinnov Optimizer technology does with its 3D microphone.
How are the listening room and its characteristics fundamental in any kind of audio application (hi-fi, pro, home cinema)?
The acoustics of the listening room is indeed the fundamental element of sound reproduction. Those who have had the chance to visit a deaf room have had the unpleasant experience of a lack of room acoustics. Room acoustics are necessary for comfort and listening pleasure. Otherwise, the analytical abilities of the brain are not fed which creates discomfort. At the same time, room acoustics can radically distort the sound. To be convinced of this, just look at the differences in reproduction when you listen to your favourite music in your bedroom, then your living room, then your car, or even at a concert. The list can be completed by studios and cinemas. The only solution is to standardise the acoustic characteristics of the listening room, either by passive processing, digital acoustic optimisation or a combination of both. Ideally, this standardization should be implemented throughout the entire audio production chain, from the studio to the final listener. This is our goal and that is why we are active simultaneously in the high-end hi-fi, home cinema, but also studio and cinema markets.
Are there any specificities for rooms intended for hi-fi, pro, home cinema? For each case, is there an ideal type of room, and if so, which one?
There are indeed differences between these environments, but I won’t make a distinction between hi-fi, pro, cinema or home cinema. More generally, there are 2 production lines. The first is the cinema production chain, which aims to reproduce the content in a cinema. The second is the music/broadcast production chain that aims to produce content for the living room. For each of these two chains, the speakers, the acoustic characteristics of the rooms and the applicable standards are specific and optimized for the application. For example, a movie studio is a high-performance miniature movie theatre and a music studio is an idealized living room. We can say that there are 2 types of ideal rooms: cinema and music/TV. Although different, these two types of ideal rooms share many characteristics:
• Direct sound, that is to say the sound emitted by the loudspeaker and reaching the listener directly, must be as perfect as possible: have a response according to a target curve, have no phase rotation, have no time lag…
• Reflections that follow the direct sound should be as small as possible because they colour the sound.
• Then the reverberation must exist (for hearing comfort) but be as neutral as possible with a reverberation time between 200 ms and 400 ms and a uniform spectral content, without any drag that privileges certain frequencies.
• Finally, resonances in the low frequencies (also called eigenmodes or standing waves) should be eliminated as much as possible.
In detail, the 2 typologies are a little different, especially in terms of the target curve, i.e. the overall frequency response to be achieved, but the general spirit is quite similar.
What services does Trinnov and its equipment bring to the different communities (hi-fi, pro, home cinema)?
Trinnov’s core business is digital acoustics, i.e. the use of digital processors to control acoustic waves. The Trinnov Optimizer acoustic optimization technology is available in all our products to meet our mission of increasing sound quality and especially the transportability of content from one environment to another, for example from a music studio to a living room. Beyond this central function, Trinnov products provide additional functions:
• For hi-fi: a 12-source preamplifier including a phono stage, a network streamer (Roon, UPnP, DLNA), a 4-channel DAC and a 4-channel ADC.
• For home cinema: a 27-source pre-amplifier including HDMI, a 32-channel DAC, an 8-channel ADC, a network streamer, a decoder/renderer/upmixer for multi-channel and immersive formats (Dolby Atmos, DTS:X and Auro-3D) and remapping to correct speaker placement errors.
• For studios: a signal analyzer (the Trinnov Smart Meter for measuring content, including perceived sound level) and a monitoring controller for managing different sources (including sub-mixes), different listening systems (speakers and headphones) and intercommunication between musicians and sound engineers.
•For cinemas: management of cinema sources (including DCI digital cinema content, alternative content) and automation.
But Trinnov’s intervention does not stop at the simple functionalities of the products, we work to develop technological expertise around Trinnov’s areas of excellence. We regularly organize training sessions for our distributors and resellers on the issues of Acoustic Optimization, immersive sound and loudspeaker placement. On this last point, Trinnov is the first company to have developed loudspeaker placement recommendations allowing optimal repro-duction of the different immersive formats (Atmos, DTS:X and Auro-3D) over a large listening area. To intensify this effort to disseminate technical expertise, we are expanding our offices and opening a new training centre.
At the beginning of your activity with Trinnov, how were you welcomed by the different communities (hi-fi, pro, home cinema)?
At the beginning of Trinnov, acoustic optimisation (sometimes called active acoustic correction, which is a bit inaccurate as some aspects are more compensated than actually corrected) was still in its infancy and certainly viewed with suspicion by the professional and sometimes even the scientific community. But thanks to the support of renowned scientists and the first pioneering and visionary clients (such as ORF Austrian Television, the BBC, NHK Japanese Television and FOX Studios) these technologies have finally become commonplace due to their undeniable efficiency and their increasingly simplified implementation. It is a real work of evangelization that has been carried out, pragmatically with the help of scientific publications, patents, but also and above all through human contact and our intense collaboration with the professional world, of which we are very proud. Today acoustic optimisation has become a necessity and not a simple palliative to passive acoustic treatment. It is an indispensable complement to deal with acoustic problems that passive acoustic treatment cannot deal with for fundamental physical reasons. We are very proud to have contributed to the emergence and democratization of this cutting-edge technology.
Are there differences in approaches to acoustics in different communities (hi-fi, pro, home cinema)?
The advantage of having a technology based on the laws of physics is that these laws are universal and apply to any environment in which one or more speakers are placed. That said, some environments have specific constraints that exploit different aspects of acoustic optimization:
• Living rooms are the least treated. It is a question of preserving the aesthetics with an acoustic treatment that is as discreet as possible (simply avoiding a reverberation time that is too long) and letting Acoustic Optimisation work.
• Dedicated home cinema rooms raise the 3 main problems:
• The presence of a screen that, although transonorous, disturbs the direct sound.
• The presence of loudspeakers with very different characteristics (screen loudspeakers are different from surround loudspeakers and at the same time different from elevation loudspeakers).
• The presence of strong resonances in the low frequencies because the rooms are generally small or medium sized and equipped with powerful subwoofers.
• Studios are in principle the most careful environments in terms of acoustic design, but acoustic optimization remains totally relevant especially if it is integrated from the design stage of the studio. But studios cover a wide range of situations, sometimes far from ideal, such as mobile studios or home studios, where acoustic optimisation is not only mandatory but simply vital.
• Cinemas raise 3 main issues:
– The presence of a screen which, although perforated, disturbs the direct sound.
– The presence of speakers with very different characteristics (screen, surround and elevation).
– The presence of reflections on the walls.
Other situations, such as the passenger compartment of a car exploit other aspects but Acoustic Optimization remains relevant in all cases.
How do you view the different communities (hi-fi, pro, home cinema)? Which ones are easier to work with?
Each of these communities has its own specificities and we benefit from working with all of them. Throughout the history of Trinnov, we take advantage of our partners’ knowledge to share it with our users in all our target markets. Our «transversal» expertise allows our home cinema customers to benefit from our technical advances and those of our partners in the professional world. It is a real strength for Trinnov to be able, for example, to offer the expertise of a mastering engineer to one of our hi-fi or home cinema customers thanks to the constant evolution of our technology.
In what terms do you think sound reproduction (in the different fields) will be modernised over the next twenty years?
Foresight is always a somewhat risky exercise. Having said that, it is possible to distinguish fashions, technological trends and invariants. What is not going to change is the physics of acoustics and the functioning of the human brain. The problem of acoustic optimization will continue to arise whenever there is sound reproduction in a space. But it is a doubly complex problem. On the one hand, there is the incredible complexity of how sound waves develop in rooms, how they propagate, reflect, diffuse, diffract in time and in the 3 dimensions of space. On the other hand, there is the even greater complexity of human hearing and especially the cognitive functioning of the brain. It is clear that there are still many mysteries to be unravelled, which can largely fuel continuous technological progress for the next 20 years, and probably more. One could imagine a kind of more intimate symbiosis between the 3 constituents of sound reproduction: the characteristics of the loudspeaker that produces the initial sound, the characteristics of the room that reacts to this initial stimulation, and the acoustic optimisation treatment that optimises the coupling between the loudspeaker and the room. As a first step in this direction, acoustic panels that integrate a loudspeaker and a microphone to boost its performance appear. Alternatively, acoustic loudspeakers are usually designed to naturally produce a flat, axially extended frequency response. This is a huge constraint that requires many compromises. Another approach would be to anticipate the fact that there will be an acoustic optimisation that will restore the frequency response and to optimise the criteria of distortion and directivity (dispersion of the loudspeaker). Still on the subject of symbiosis, I think that we can reasonably bet on the reinforcement of audio integration and the principle that the listening room, its architecture, its acoustics and its audio system will be designed and integrated together. Beyond the invariants of physics and technological progress, there will also be fashions, new content, new formats, new ways of accessing content, consumption habits, design fashions,… Here we are entering a much more volatile and delicate field to anticipate over 20 years.
How can technologies such as active digital filtering improve the performance of systems?
In a loudspeaker cabinet, the frequency band must be adapted to the capabilities of each speaker, for example, bass, midrange and treble. Most hifi loudspeakers use passive filters with passive power components (resistors, inductors and resistors) that introduce insertion losses in the signal path. For example, if the tweeter is 3dB more sensitive than the woofer, the resistor used to attenuate the 3dB tweeter will dissipate half of the power delivered by the amplifier. In addition, the passive crossover introduces a degree of freedom between the amplifier and the loudspeaker which limits the control of the latter. The active crossover solves this problem by driving each loudspeaker with a dedicated amplifier ensuring the best possible efficiency and control, the frequency band of the signal being adapted to each loudspeaker upstream of the amplifier. This approach is preferred in the studio. With its Optimizer technology, Trinnov takes the active filter concept to a new level of performance. The active crossover is adapted not only to adapt the frequency band, but also to take into account the acoustic characteristics of the loudspeakers and the listening room in order to ensure the best possible connection between the channels at the listening position.
The acoustic optimisation of the active crossover also controls the directivity of the loudspeakers at crossover frequencies. It ensures that at each crossover frequency, the power emitted by the loudspeaker is concentrated mainly on the listener and not on the walls, so that the result is less dependent on the listening room. It is a kind of acoustic optimization at the source. For example, we work with several manufacturers of high-end loudspeakers who rely totally on our processors for the part dedicated to filtering and concentrate on their core business, the electroacoustic part. We can mention Vidid Audio, Dynaudio, B&W or even AudioNec or Bohne audio in Germany.
What is today the Trinnov machine park installed in the different communities (hi-fi, pro, home cinema)?
We have equipped more than 4.000 high-performance audio installations worldwide, including more than 1.000 studios and more than 1.200 cinemas.
On a purely technical level, what room for improvement do you have? What are your future directions?
As I mentioned earlier, acoustic optimisa-tion and more generally digital acoustics are new fields, relatively little studied and particularly complex: the way in which the waves emitted by a loudspeaker propagate in 3D in a listening room is complex and the way in which a listener perceives them is even more so. So there is still a lot of room for improvement. By way of comparison, the electrodynamic loudspeaker we know today with its diaphragm, coil and magnet was introduced on the market in 1920 and continues to be perfected to this day.We are therefore in a logic of continuous evolution, crossing scientific and technical work with the field experience gathered by the 4.000 high-performance installations using our Optimizer technology, including more than 1.000 of the most renowned studios in the world. To talk more concretely about future directions, we are working, for example, on improving bass reproduction by subwoofers. If the work is conclusive, this will lead to an improved version of the algorithm.
