Audyssey Sees Half Its Revenue Coming from Streamed Music in 3 Years

Bandwidth isn’t the leading cause of poor sound quality in the mobile streaming environment, said Chris Kyriakakis, Audyssey’s founder and chief technology officer. Instead, he said, it’s the acoustical limitations of the headphones, and Audyssey hopes to capitalize on reducing those limitations through the AmpIT software development kit it released this week for streaming music providers. Kyriakakis cited the tiny transducers behind plastic that don’t seal with the ears and cause leakage, which reduces bass, along with resonances in the sound cavity and “all the acoustical problems you can think of in a cheap speaker.” While acknowledging that higher bandwidth generally translates to higher quality, Kyriakakis told us, “After a certain point, in order to hear this improvement from better bandwidth, you'd better fix the sound quality of your playback device. Otherwise you're wasting your bandwidth.”

Audyssey has released the AmpIT SDK for streaming music providers to complete the software/hardware solution it announced with its ExpertFit headphone project last month (CED May 13 p3). ExpertFit allows headphone makers to automatically fix acoustical problems of listening devices, allowing users to select the profile for a particular headphone model from the Audyssey website. The AmpIT app is the processing engine inside a music streaming app that plays the sound according to a headphone’s profile. Kyriakakis called AmpIT a highly optimized piece of code “with Audyssey inside,” and the company is pitching the engine to “everybody who has an application that’s streaming music."

There are no technical requirements on the streaming side to be able to use the AmpIT software, Kyriakakis said. The software is “completely independent of data rate,” he said. The code is a “simple drop in” to the iOS platform, Kyriakakis said, and on the Android side, it’s only compatible with Android 4.x and higher. Earlier Android platforms “weren’t really built around media processing in general and had a very hard time processing media,” he said.

The licensing model for the technology is volume-based, Kyriakakis said. To date, Audyssey’s licensing revenue has come from the CE hardware side where its audio processing software is used in AV receivers, TVs, set-top boxes, theaters, cars with digital amplifiers with DSP chips. The formula has to change for streaming music companies who reach millions of listeners across various devices, and Kyriakakis called it a “sliding scale” but declined to specify fees.

"We have to adapt to this changing world,” Kyriakakis said. Before, to run any kind of processing, Audyssey needed a box “that had a chip in it that could run stuff,” he said. The smartphone and tablet are now the boxes with the processing power built in, “and we started looking at how to shift the block diagram so that the Audyssey stuff doesn’t have to be right where the listening device is but further up the stream,” he said. That led to optimizing the software to run at the application layer in a way that doesn’t require app developers to have to do anything other than plug in the code, he said.

For the first time, Audyssey is venturing into markets that aren’t known for having processing logistics, such as headphones. “It’s a big deal for us,” Kyriakakis said. Going forward, “we want Audyssey to be the backbone of sound, one device at a time,” he said. The company is banking on smartphone device makers wanting to differentiate one handset from another with better sound quality for music. Some smartphone makers are evaluating the technology now to determine how to make more advanced media players than the generic ones on the market today. “That’s where we would fit,” he said. Three years from now, Audyssey expects revenue share from the mobile side to account for more than 50 percent of revenue. “That’s a big shift,” he said, “but it’s following the shift in the market."

To establish a reference point for audio quality, Audyssey went to several mixing studios representative of different genres and measured what audio engineers were hearing in the environment where content is being created, Kyriakakis said. Studio engineers listen to loudspeakers in a calibrated well-designed room “sometimes with Audyssey, sometimes without,” and the sound that enters the ear is the reference, he said. “Whether we like it or not, that’s what they made,” he said. Audyssey engineers use the company’s MultEQ technology to determine the differences between what is heard by audio engineers at the mixing console and what is heard through headphones by placing miniature microphones inside an ear canal simulator positioned in a mannequin, Kyriakakis said. “We found a way to translate the response of a good calibrated studio at the mixing position to inside the ear canal,” Kyriakakis said. “It’s a direct path back to the original sound,” he said.

Every headphone in the company’s database has been measured on the mannequin, Kyriakakis said. Audyssey studies the differences between the headphone and the studio sound and creates a filter, or a profile, that goes to the cloud-based database. When users install the Audyssey app and plug in headphones, the app shows a list of headphones, and users choose the one for their headphone model. The database currently holds 205 models from 55 manufacturers, he said. He estimated the 200 models cover 80 percent of the headphones in the U.S. market. Audyssey has planted a suggestion mode inside the app that consumers can tap into, if the app developer enables the feature, so users can suggest models for the database and then receive an email if it’s been added to the list.