New meeting kits from Google provide better audio using local AI

Meetings can sometimes be frustrating for remote participants, who often have to contend with extraneous noise, unclear speech, and bad acoustics, especially with multiple participants in large conference rooms.

High-end conference room setups can alleviate these issues with individual mics for participants and professional audio equipment such as high-end digital signal processing hardware and software. But these can be expensive and cumbersome for most businesses to deploy and maintain.

Google’s new Series One meeting room kits, in partnership with Lenovo, allow for businesses of any size to create spaces that filter out unwanted noise while focusing audio pickup on individual participants. The new kits use Coral’s local AI acceleration to enable on-the-fly noise removal for clear audio, from each meeting participant.

Clear communications for better meetings

“Our north star is to enable the clearest possible communication in a challenging environment,” says TJ Varghese, Group Product Manager for Google Meet hardware. And he doesn’t just mean unusually bad environments to host meetings, such as lunchrooms doubling as conference rooms. Any conference room can present challenging acoustics that can stymie comprehension for remote participants.

The new Series One meeting room kits add local AI-enabled functionality from Coral to this Google Meet hardware lineup to let meetings come through loud and clear to those conferencing in.The secret to clarity without expensive and cumbersome equipment is virtual audio channels and on-device audio processing.

The Series One Smart Audio Bar incorporates internal microphones that together can create separate beams — that is, directional audio channels — that provide focused attention on individual speakers. Adding up to four optional mic pods — each with eight channels — allows for easy expansion.

“By isolating that sound, you create a very high signal-to-noise ratio between what’s in front of you versus what’s 20 degrees offset from you,” Milton Ribeiro, Hardware Lead for Meet Meet hardware at Google, adds.

In other words, the Audio Bar creates virtual channels that effectively act as individual microphones for each meeting participant, without requiring them to mic up or fiddle with tabletop microphones. The effect is to create the feeling that even remote participants are present with those physically in the conference room.

Cutting through the noise

Beyond merely isolating individual speakers for maximum clarity, the system goes a step further by eliminating unwanted sounds from the audio stream going out to remote participants.

Coral’s Edge TPU enables multi-channel noise cancellation technology called TrueVoice® —to let every voice come through loud and clear—while filtering out distracting sounds, such as snack wrappers, sirens, and typing.

The Edge TPU coprocessor allows the system to process audio locally as the system cannot rely on cloud processing. That’s because the large number of channels the system supports (up to 44, including channels provided by the mic pods) would create unacceptable lag and make conversations jumpy. “You can get similar performance at much higher cost and much higher power using a GPU, for example, but we are very power-limited,” Ribeiro explains.

Coral’s low power processing is ideally suited to meetings, Varghese says, for two reasons. First, the Coral chip’s low wattage doesn’t produce enough heat to require fans for cooling. No fans mean no extra noise to clutter up meeting audio. Second, low-power processing enables the new system’s minimalist, hassle-free setup. It runs on power over Ethernet, or PoE, reducing the number of cords and freeing up power outlets.

Both the Series One Smart Audio Bar and Meet Compute System feature the Edge TPU. In the Compute System, Coral’s new M.2 Accelerator with Dual Edge TPUs provides double the number of inferences per second as a single TPU, powering applications at a blistering eight trillion operations per second, using a total of only four watts of power. The new low-power powerhouse is great for vision processing applications and will enable an expanding array of capabilities for Series One Meet hardware.

Like the Chromebox, the new Smart Audio Bar also includes dual Edge TPU Accelerators soldered directly on board. Future-proofing is the name of the game here, as a single Edge TPU would do the job of isolating audio from individual meeting participants and filtering out noise. Including additional accelerators from the get-go will enable future capabilities, for example by allowing more than one machine learning model to run at the same time.

Powering meetings of the future

As for future functionality the Edge TPU might enable, the Google Meet hardware team is keeping its plans closely guarded. But Varghese does say local AI can do much more with sound detection and processing. And announced plans for Series One future updates making use of conventional CPUs provide a hint of the kinds of enhancements that local AI could bring. These include a magic frame, generated on the fly for meeting cameras, that focuses on the participant, making it easier to follow along with the speaker, even if they’re moving around the room. It seems the team is just getting started making remote meetings better for all.