Interview with Eduardo García
by Markus Thiel – June 2020
The producers of German Wahnsinn prove in exemplary fashion that audiobook productions can be much more than the well-read recording of a printed bestseller. Their current audio translations of the written word have meanwhile been published by their own publishing house, founded together with bestselling author Cornelia Funke. And the name of the label, "Atmende Bücher" (Breathing Books), which is currently becoming increasingly successful, certainly delivers on its promise. Against the background of the production "Das Labyrinth des Fauns" ("The Labyrinth of the Faun"), which was created in cooperation with and based on a draft by Guillermo del Toro, we spoke with German Wahnsinn founder Eduardo García about the transformation and further development of the classic audiobook, as well as about recording philosophies in the face of modern immersive audio concepts.
As far as I have heard, you are not only very active in the audiobook sector, you have even founded your own publishing house. Can you tell us something about that?
I am actually one of the very first people in Germany who produced audiobooks in the early ’90s during the first wave. I met a producer back then who had imported the whole concept from England — the format was called "talking books" if I remember correctly — and then asked me if we could produce something like that in my studio. That's how the first series was created, with narrators like Uli Wickert, who was still a newsreader at the time, or illustrious guests like Joachim Witt ("Der goldene Reiter" ed.). At that time, of course, we still recorded everything on 1/4-inch tape. When the real audiobook boom began, I was already out again — which was a real shame in hindsight! But my passion for it never really died. Later on, I worked as a service provider for larger publishing houses for a long time, as they gradually added an audio division to their portfolio. At some point I received an inquiry as to whether we could compose and produce the music for an audiobook by Cornelia Funke. This was about the "Reckless" series, the fourth volume of which will be released in November 2020. In the end, there was a lot of gloomy cello music and after the audiobook was released, we even had public readings where we could perform the music live. This was also what led to us getting to know Cornelia a little bit better and, what’s more, the period when we received the nicest compliment of all time; she came up to us after a performance and told us that when she first heard our music, she already had the feeling that we had understood her book better than she had. Since then this has developed into a real friendship and to us wanting to produce more audiobooks together in the future. This is what we are investing a lot of time and effort into doing now and, more recently, also from our own publishing house called "Atmende Bücher".
How did you approach the first joint production?
The first audiobook production was already very complex, comparable to a film score — but then the whole thing took almost 9 hours! The last production we did together was the audiobook version of the film "Pan's Labyrinth" in collaboration with Guillermo del Toro. On "Das Labyrinth des Fauns", which is the book title, we finally decided on a consistent 3D audio rendering, together with Cornelia, who wrote the audiobook version and who then enhanced it in conjunction with Guillermo. Since we currently work a great deal with VR and VA anyway, it was the logical next step for us to try it with an audiobook. In the end, we had to start from scratch four or five times until we had got the dramatic composition right. That was a really important process in terms of the final binaural mix.
Did you record binaurally straight away during the production?
Yes, we also have special microphones for this. Cornelia also narrates a few parts, which she then did directly around the 3Dio microphone set. That way you get a bit of movement in the end. We even recorded most of the Foley sounds in 3D in advance and modulated some of them with Nuendo later on. The real challenge here is that you can't really get narrated speech as up-front in the mix in 3D as you can in stereo. Our goal is to get the whole thing to work in such a way that the listener enjoys it. This was why we had people who were not involved listen to the unfinished recordings time and again, and had to take a lot of criticism, especially from purists. In comparison to a stereo mix, 3D voice only achieves a maximum of 90 percent of the usual up-front factor, but once you really get into the mix you suddenly feel like you’re right in the middle of an audio film.
That’s exactly the impression I got after a first listen. It may be unfamiliar at first, but in the end it's much more intimate due to the way it’s mixed.
Exactly, because we continually place the narrator in different acoustic spaces, you, as a listener, completely forget at some point that it’s a narrator. Of course, you could take issue with that and say you could make a radio play out of it in this way, but the narrative form is the audiobook! But that's exactly what made it so interesting for us. All in all, the production process really took a lot of time, I mean, we spent the first half hour redesigning the whole story five or six times over, including the music, until we finally had the feeling that you can really immerse yourself in the way we imagined it at the beginning. With the German version, we have now hit over ten million streams on the digital platforms, which just goes to show that we managed to get the whole thing right in the end. Of course, taking such an approach also carries an enormous risk, because in the end, the time and effort we put into making our nine-hour audiobook version is, in terms of production, about the same as what you would put into making three whole films.
I can well imagine. How much time does a production like this ultimately take up?
At some point I added it all up and came up with a production time of two and a half to three months — and that was with six of us working on it in total. That was a very special experience for everyone involved. The interesting thing for me was that in doing "Das Labyrinth des Fauns" we, along with Cornelia, took it upon ourselves for the first time to create an audiobook designed for the adult world, i.e. from the age of 15 upwards. The previous productions mainly had a much younger target audience, so this time we were able to do some darker stuff here and there.
But you are currently taking things a step further in the area of AR?
We've been working for a long time on how to get even more immersion into the actual listening experience. We've even developed an app that works together with the new accelerometer and gyroscope equipped headphones and glasses, for example, and allows you to influence the course of the radio play by gestures such as nodding or shaking your head — in other words, the integration of an active decision-making process. We have also adapted the whole thing to the gyro sensors of a smartphone, which ultimately makes five to six different decision trees available within a story. Which in itself is a pretty crazy thing, because if you follow this concept, a radio play that is only 15 minutes long will suddenly become endless to produce! The current demo project is based on a spinoff that Cornelia wrote last year and which we have enhanced and redesigned a bit for this purpose in order to sound people out in the scene as to how far our app could be interesting as a flexible foundation, since companies like Apple, but also platforms such as Audible and Spotify, Audioboom with its Originals podcasts, radio plays and many more, are making productions more and more perceptible. This technology is also particularly interesting for advertisers, who we also deal with. By cleverly switching between stereo and 3D audio you can catch peoples’ attention in a very focused way. First of all, we develop these things exclusively in Nuendo before transferring everything to the appropriate middleware option such as Wwise, FMOD or Unity.
Do you set the direction, panning etc. directly in Nuendo?
We really do a lot of things directly in Nuendo. Apart from pans, the tricky thing is to find the right acoustic spaces, a discipline which many competing systems unfortunately completely fail at. But it is, and remains, a lot of fiddling around, where you always get to points, where existing things have to be combined and reassembled. We also have to be flexible towards the developers’ wishes — one works with Wwise, the next one can't use it at all and swears by FMOD, and then another does everything in Unity. In principle, we are then the actual middleware that always has to keep an eye on the fastest and most optimal workflow and the most professional implementation options, measured against the desired requirements. The 3D audio stories are actually still in their infancy, but fortunately we are also noticing that on that issue we are slowly moving closer together in terms of workflow. Little by little everyone just gathers more experience. The threshold which we are currently at can be compared to the transition from mono to stereo. A completely new generation is currently growing up and entering into immersive audio playback processes.
In a sense, we’ve left behind the experimental generation of dummy heads, which unfortunately did not really come into its own at the time, since many further processes and implementation possibilities didn’t even exist.
Yes, today everything is finally mixable with everything else. In addition, it is also available to almost everyone, which means that everyone can put everything out the way they want it. There are just some really great and revolutionary things and, of course, a couple of not so beautiful things (laughs.) On the other hand, you also learn something new every day — especially in our team of senior and junior engineers, which now consists of 12 staff. The younger colleagues, who have grown up with it as a matter of course on their way to reaching the goal of using dozens of different software tools, provide us with a lot of important input. But, beyond the boundaries of your own company, I think it’s the interaction in general that makes this sector of the audio industry so exciting. The passion for a cause is actually the most important thing for us.
What are the next projects you’ve got planned?
Quite a few! We have just finished a makeover of the first part of Cornelia's "Drachenreiter" (Dragon Rider) series. For this purpose, we have acquired the rights to the sixteen-year-old recordings from the original publisher, which we are now editing in the style of the second part, which we already produced. Next year the third part will be released — then the trilogy will be complete. Until October, we’re producing the fourth part of Cornelia's "Reckless" series, which will be released as an audiobook parallel to the book release in November. So, we’ve still got an awful lot going on this year — and these are just the fun projects! We’re also producing audiobook series such as “The School of Magical Animals” or “My Friend Conni” that run parallel to our daily studio work and the many other jobs like Nivis World, a Nivea storytelling app, or the the audio for a VR simulator application. It never gets boring here and, of course, we make sure that everyone is still having fun at work — because that's the most important thing for us.