SRS Labs Pushing Object-Based Audio at CES 2012

SRS Labs is usually known for its surround-sound emulation algorithms. But the company now has a bigger idea: revamping the entire way the audio industry produces sound.

This could be much larger than an SRS WOW HD button, so it's worth discussing. At CES 2012, SRS Labs is pushing the idea of multi-dimensional audio, which focuses on audio in terms of objects, instead of in channels (such as two-channel stereo, or 5.1-channel surround sound). Rather than mixing individual instrument tracks in a song, or mixing ambient sound, sound effects, and dialog in a movie's audio track, the engineer instead takes those audio pieces and directs exactly where they go in the listener's physical speaker configuration, as well as how loud they play.

In other words, instead of an engineer producing a finished, static mix that plays back the same way regardless of how the playback system is setup—and if the playback system isn't any good or set up incorrectly, tough luck—the engineer produces a finished bundle of meta-data, complete with digital instructions on where and how all of the audio pieces play. Then an MDA-compatible renderer, either in software or built into consumer electronics components, decodes it properly for the listener's playback system.

This is a subtle but key difference. "For example, once we create this way, we can do an audio program mix on 11 speakers," said John Kellogg, executive director of corporate strategy at SRS Labs, in an interview. "Think of a 7.1 speaker with four more speakers for height. Now we can take that mix in a dubbing stage on 11 speakers and the MDA player or renderer, as we call it, can map it to any number of speakers the consumer has. It's one deliverable that translates into any environment."

The consumer programs the player in the very beginning: "Here's how many speakers I have in my room: two, five, seven, or whatever; here's where they are, here's how far away from me they are," Kellogg continued. "Once those coordinates are in, the MDA player maps that audio program beautifully." The goal on the low end is to help people with two speakers, or a soundbar, to get a much better, 3D-like audio experience than they're getting now, and to help people who didn't place their speakers properly. On the high end, the surround sound experience would be more defined and transparent than it is now. SRS Labs talked a lot about systems with 11 speakers, 22 speakers, and so on with us, but we see that as mainly for commercial installations; we're not expecting the average home theater owner to go for that sort of thing.

Many of us have heard amazing-sounding, THX-certified surround systems over the past two decades. With the right setup, you could achieve excellent surround sound even in the analog Dolby Pro Logic days, which dates back to the 1970s and dominated home theaters in the early 1990s. So I pressed for more details on why MDA is different, and proposed the example of an actor running across a screen shooting bullets. I received this response from an SRS Labs spokesperson; pardon the two block quotes, but it's important to use their words exactly here (edited slightly for clarity):

"Let's say said action star runs across the screen shooting bullets—since MDA stores the spatial data for specific sounds and for how they travel throughout the action sequence, all aspects of the action sequence would feel as if it were actually taking place all around you. You would hear the ambient noise in front, to the side and behind you, and bullets would sound like they are whizzing past your head (from in front of you past you). You would be able to just close your eyes and feel like you can reach out and point to the exact location a sound object is coming from. In a sense the action sequence would come alive in your living room or in the cinema."

Fine, but again, how is that different than what we can already do now?

"Regardless of how great a current home theater or movie theater set up is, the fact is, the source content is still mixed for planar (2D) reproduction. MDA feeds a home theater with true 3D audio data, and it is the playback device's job to render that 3D audio data into an accurate representation of the original mix to the best of its capabilities. Having 3D positional sound data allows an MDA playback device to utilize all of the device's speakers, plus a variety of filters and signal processing solutions on the fly to make each individual sound of a mix come to life, enabling sounds to literally be peeled off the walls and thrown out into the sound stage to a pre-determined location."

In other words, today's movie and music CDs play back the same no matter what the playback system is. Sure, some home receivers let you set it to 2.0 or 5.1, but either way, the source material (the movie) is the same in all cases. An MDA audio renderer changes the way the audio is mixed depending on the room configuration. It puts an unprecedented level of control in the audio engineer's hands, and can deliver 3D-like effects using just two speakers or a soundbar in a way that you could never get from an album or movie mixed down to stereo (at least, that's the claim).

Essentially, SRS Labs is taking the adaptive, object-based audio that the game industry uses (such as Microsoft XACT, or Creative Labs ISACT), and putting it into a form that works for linear audio, but with different speaker configurations. Instead of converting sound effects to objects and placing them in a 3-D field—say, for an enemy in Call of Duty: Modern Warfare 3 with an automatic weapon, shooting as he runs around in front of you—SRS's new system lets audio engineers take apart the pieces of a soundscape and place them in the appropriate speakers in the listener's own physical environment. It's an existing idea, but applied to a different kind of situation. But it also means you need a movie or music album mixed in NDA and an MDA-compatible system to decode it correctly, so it has to be a new industry standard across the board.

Things like this get the audio community excited on a periodic basis. SRS Labs said some big names are on already board for MDA, including Skywalker Sound. But there's a huge difference between that, however, and actually putting MDA-compatible material and MDA-compatible products in front of consumers—most of whom have already flatly rejected audio-enhancing technologies like SACD and DVD Audio, and many of whom don't bother to wire up the rear surround speakers that come with a home theater system.

SRS Labs is hoping the industry comes together and joins on a standard the way they did with MIDI. But that happened in 1983. We've never seen anything like that on MIDI's scale since. I also have my doubts that consumers will put in the effort to measure out their audio system in such detail. If someone is patient enough to program an MDA-compatible product with a detailed speaker configuration, that person is also patient enough to place quality speakers properly in a traditional 2.1 or 5.1 system to begin with. That same enthusiast may thrill at the level of realism MDA provides, but could MDA capture the mainstream consumer's interest? Would it matter to that person if a soundbar was "smart," or MDA-aware, in this way?

There's no word yet on actual MDA-compatible products or source material yet, but SRS Labs claims to have a lot in the works. Whatever happens, we're for anything that improves audio playback, as long as it actually improves it in real-world situations. Here's hoping the company is onto something that transcends yet another pseudo-surround sound enhancement.