Smart speaker development platform leverages AI
Built to deliver evolved premium audio experiences throughout the home, the QCS400 SoC series, says the company, help deliver highly optimized, AI-enabled solutions that are designed for smarter audio and IoT applications. They integrate high-performance processing, the Qualcomm Artificial Intelligence (AI Engine), connectivity, and multiple advanced audio and visual display capabilities in a single, power-optimized chip architecture.
In addition, says the company, for users of smart speakers, the SoCs support a more robust voice assistant experience with faster, smarter voice UI, even in noisy environments compared to the company’s previous SoCs. Features include AI-based local automatic speech recognition, low-power, multi-keyword far-field voice pickup with beam-forming and echo cancellation, and support for cloud-based voice assistants.
“These new SoCs raise the bar on both feature integration and power performance for smart audio compared to our previous technology,” says Rahul Patel, senior vice president and general manager, connectivity, Qualcomm Technologies, Inc. “This will help manufactures to more easily overcome significant technical challenges and build smarter speakers and assistants with more intuitive voice UI, connected user experiences and exceptional sound quality.
“The next generation of smart audio products must be robust, highly interoperable, feature-rich, and smart, yet extremely power efficient. Qualcomm Technologies helps to meet these needs with our new SoCs, which are unique and comprehensively integrated to combine enhanced compute, AI acceleration, and low-latency audio distribution in a single chipset. These integrated solutions are designed to reduce development time of next-generation smart speakers, home assistants, and soundbars.”
Key features of the QCS400 SoC Series are offered as the following:
- Highly-integrated, yet flexible, single-chip architecture – with the Qualcomm AI Engine, dual DSPs and up to four processor cores, engineered specifically for audio applications and designed to deliver substantial power efficiencies, performance improvements and flexibility for advanced feature innovation.
- On-board connectivity – industry-leading Wi-Fi and Bluetooth Support for Qualcomm aptX Adaptive designed to deliver low-latency audio streaming in the room or throughout the home, and tightly synchronized audio and visual content for a superior entertainment experience. Wi-Fi, Bluetooth and Zigbee coexistence is designed to allow for control of virtually all types of smart home devices, helping OEMs to design products to serve as hubs and interfaces to the smart home.
- Truly rich and immersive audio – up-to 32 channels of integrated audio processing, support for Dolby Atmos immersive audio and DTS:X as well as compatibility with Qualcomm DDFA amplifier technology and aptX audio.
- Improved Voice UI – Utilizing the company’s far field voice, multi-channel echo cancellation, and high performance, low-power multi-keyword detection algorithms (Qualcomm Voice Assist), products built on the QCS400 SoCs can capture voice commands even during high-volume audio playback or from further afield.
- AI Engine – To meet the growing demands of audio processing, QCS400 SoCs support the Qualcomm AI Engine for enabling high-performance, energy efficient AI acceleration including a quad-core CPU, Qualcomm Adreno GPU and Qualcomm Hexagon DSP with Hexagon Vector eXtentions. The Qualcomm AI Engine offers support for myriad neural network frameworks including Tensorflow, PyTorch and ONNX, as well as the Qualcomm Neural Processing SDK for easily executing AI algorithms on device. This advanced AI hardware and software suite helps smart speaker and sound bar OEMs deliver a variety of robust AI-powered voice UI features.
- Reduced development time – Designed specifically to support reduced development time and cost, the new chipsets will be available at multiple tiers with a common architecture, allowing implementation across multiple product lines, to better utilize development resources.
The QSC400 SoCs are customizable using the company’s Audio Development Kit (ADK) and the Qualcomm AI Engine can be utilized by developers through the Qualcomm Snapdragon Neural Processing Engine (SNPE) Software Development Kit (SDK). Developers have the freedom to choose the cloud-based voice assistant that best suits their product vision.
To help streamline the process for developing a smart speaker, the company is also announcing the Qualcomm Smart Audio Platform 400, available as a comprehensive, customizable development kit and example design, built around the QCS400 SoCs.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
