OnQ Blog

With the Qualcomm QCS400 SoCs, immersive audio meets on-device AI [video]

The future of smart audio is here.

With the new Qualcomm QCS400 SoC series, we’re bringing immersive sound quality together with on-device artificial intelligence (AI) designed to deliver evolved audio experiences throughout the home. From the simple smart assistant to the hi-fidelity soundbar, this exciting new technology is engineered to deliver better sound quality, optimal battery life, and next-gen voice control compared to our previous technology.

We chatted with Rob Saunders, director of product marketing at Qualcomm Technologies International, Ltd. to find out how the Qualcomm QCS400 SoCs have been engineered specifically for consumer audio within the expanding IoT and how they’ll support their customers and enrich the audio experiences of consumer electronics users. 

What are the three most important user experiences that the Qualcomm QCS400 SoCs make possible?

First, the Qualcomm QCS400 SoCs support the evolutionary improvements in voice UI that are critical to realizing the vision of the connected smart home. A combination of highly-integrated tech — including our advanced Qualcomm AI Engine, far-field multi-keyword voice pickup with beamforming, and low-power voice detection — compared to our previous technology, it supports faster response times for voice commands, more robust voice pickup when people are using their speaker in noisy environments, and local voice control, which makes it possible for smart speakers to respond even when they aren’t connected to the internet.

Added to that, these SoCs can drive smarter, more immersive sound quality, particularly for home cinema experiences compared to our previous SoCs. With its high-performance, low-power architecture, the Qualcomm QCS400 can support rich audio features, including Dolby Atmos and DTS:X, that help people to have a powerful surround-sound experience with a soundbar and networked speaker set-up. People want to be further immersed in the thrill of a car chase or the excitement of a live stadium concert, and this helps to make that a reality.

The third thing to call out is that the Qualcomm QCS400 SoCs are designed to make it easier for people to enjoy this great sounding audio virtually everywhere. There are two ways we look to achieve this. One is with robust, multi-channel connectivity that is engineered to enable smart speakers to be networked together in a room or around the home. This means people can have ample control of the audio devices on their home network and play back music and audio in different rooms or the whole home at the same time. The second way we help enable great audio is by driving improvements in power performance and battery life — people can take their speakers with them into the yard or even to the park or beach and enjoy playback with voice commands.

Would you tell us more about the superior audio features you mentioned?

The new SoCs have a quad-core processor as well as a dedicated audio DSP offering up to 32 channels of processing, as well as rich features like Dolby Atmos and DTS:X. It’s all about helping our customers to deliver a rich, more immersive, cinema-like experience in the home.

In addition, with Qualcomm DDFA amplifier technology for premium audio amplification and Qualcomm aptX Bluetooth technology, we’re supporting great sound quality on other wireless audio devices like smart speakers. And, the Qualcomm QCS400 SoCs are designed to also deliver low-latency whole-home networking for virtually seamless audio streaming minus the lag or delay.

Apart from low-latency streaming, there are other rich connectivity features on these SoCs — for example, support for tri-radio coexistence of Wi-Fi, Bluetooth, and Zigbee channels all at once. This is a complex feature that supports music streaming and home control at the same time, and by having it pre-integrated we’re helping save development time for our customers. With all this combined, we believe this SoC series is really delivering quite an impressive array of audio capabilities and features.

One of the most notable advances on the Qualcomm QCS400 SoCs is the on-device AI. Can you tell us more about that?

Our Qualcomm AI Engine is an exciting feature of the Qualcomm QCS400 SoCs. AI is a growing force in all areas of technology because it gives devices the ability to learn ways of doing things, instead of relying on programmed logic by learning in this way, they can be more efficient and effective. On Qualcomm QCS400, we’ve taken this AI capability and applied it to help us make a smarter, more robust Voice UI than our previous technology. We believe that smart speakers with this technology can more accurately recognize wake words, respond faster, and perform many commands without connecting to the cloud. This is better for on-the-go use and at home control.

Our Qualcomm AI Engine is compatible both with third-party neural networking frameworks and our own Qualcomm Neural Processing Engine (NPE) SDK tools that will help developers utilize this advanced AI to create their own machine-learned features. This is a powerful capability.

How does the Qualcomm QCS400 SoC series span across different types and tiers of devices?

Qualcomm QCS400 is designed to underpin smart audio devices from a home assistant or portable smart speaker through to a high-end audio/visual receiver (AVR). And within those categories, it’s possible to have a broad spectrum of features and price points.

We looked to achieve this scalability by creating SoC variants at different tiers to suit the needs of different product categories. By including our Audio Development Kit (ADK), we also looked to provide flexibility in the SoC design to give developers the power to configure and customize a broad range of features for their end products. The Qualcomm AI Engine is also designed to be highly flexible, and through our Qualcomm NPE SDK, audio manufacturers can utilize this powerful system to create and run their own machine-learned inferences.

What does the future of smart audio look like?

It will be connected — both to the cloud and between local devices. This means robust networking across all channels will be a critical technology requirement.

It could be even smarter, with more intuitive and robust voice interaction and more opportunities for personalization (through biometrics, for example) and security features. These things are necessary to help truly deliver the convenience benefits that smart audio devices could offer people.

Finally, sound could be even higher quality and more immersive across audio devices. From wireless headphones and earbuds, to soundbars and home cinema. Our 2018 State of Play report shows sound quality has repeatedly been the highest-ranked factor by consumers when choosing wireless audio products. That’s not going away, and I think people will become even more aware of sound quality and care even more about it over time.

Qualcomm QCS400, Qualcomm AI Engine, Qualcomm DDFA, Qualcomm aptX and Qualcomm Neural Processing SDK are products of Qualcomm Technologies, Inc. and/or its subsidiaries.

Opinions expressed in the content posted here are the personal opinions of the original authors, and do not necessarily reflect those of Qualcomm Incorporated or its subsidiaries ("Qualcomm"). Qualcomm products mentioned within this post are offered by Qualcomm Technologies, Inc. and/or its subsidiaries. The content is provided for informational purposes only and is not meant to be an endorsement or representation by Qualcomm or any other party. This site may also provide links or references to non-Qualcomm sites and resources. Qualcomm makes no representations, warranties, or other commitments whatsoever about any non-Qualcomm sites or third-party resources that may be referenced, accessible from, or linked to this site.