Audience earSmart eS325 Voice Processor

We’ve talked about Audience’s eS515 sound processor during CES, and here at Mobile World Congress (MWC), Audience is releasing the es325, which is a version that mostly brings the same feature set (more on that in a minute), except that it is targeted at customers (device makers) who want to integrate Audience’s products into an existing audio system (hardware and software), while the eS515 is more of an “all in one” solution for those who want an end-to-end solution. In practice, this means that more devices are going to use Audience’s sound processing, including PC manufacturers. If you are unfamiliar with this type of chips, the bottom-line is that they process the sound (incoming or outgoing) and make voice clearer by removing everything that’s identified as being background noise.

I tested it with the Audience team over a Skype call (they were using a laptop equipped with the es325), in an environment where there was a copious amount of background noise (restaurant-like crowd). The result was very impressive: most of the crowd noise was toned down to the point where it was barely distinguishable, and the voice of my call recipient was very clear – much more so than with any other audio system I had tested before.

The same chip can be used in smartphones or tablets, and given that just about everyone cares about in-call audio quality, there is clearly a high demand for this type of product. Unfortunately for Audience, there is no “audio quality” benchmarks like there are “performance” benchmarks so most people are unaware of the vast potential improvements that audio processing can bring, but they are working on several ways to demonstrate their products. In the meantime, there’s nothing like a live demo.

The technological background for doing this type of audio processing is complex, but audience is using several key techniques, like using three microphones simultaneously (most systems only use 2 out of 3 when all three are available), which gives the Audience processor a notion of “directionality” that the human ear has (because of how it is shaped). We “humans” tend to better hear (and pay attention to) sounds when they are coming from the front. Microphones work in a completely different way, and tend to pick up the sound from any point around them, that’s why there are “microphone-guns” which are optimized for directionality. With three regular omni-directional microphones, Audience is capable of extracting the direction from which the sound is coming, and knowing “where to listen” is key in ignoring unwanted sounds. After all, that’s what our brains do all the time.

We expect the company to announce partner products in the near future, but previous-generation Audience design-wins include high-profile products like the Samsung Galaxy Note 2, just to cite that one. With an entry in the laptop world, the odds that your conversation will be powered by Audience just shot up.

Hubert Nguyen
Categories: Audio
Tags: MWC, MWC 2013, Processors, voice