Vendor: Ceva, Inc. Category: Edge AI Accelerator

Neural-network-based noise cancellation

In a world where videoconferencing, team gaming, and voice-operated systems proliferate, it is vital to extract clear, intelligib…

Contact Provider Request Datasheet

Overview

In a world where videoconferencing, team gaming, and voice-operated systems proliferate, it is vital to extract clear, intelligible voice from environmental noise. Ceva-ClearVox ENC software features a neural-network-based ENC algorithm with small memory and processing requirements to embed ENC into even tiny systems. The algorithm processes both outgoing and incoming speech to provide clear communications to both parties on a call

The Solution

Ceva-ClearVox ENC implements a trained neural network in embedded software, eliminating the need for external cloud connectivity. Despite the neural-network approach, ClearVox ENC requires so little memory and processing power that it can be coresident with other applications even in tiny hardware configurations. Yet the algorithm is effective enough to give excellent speech quality for videoconferencing and gaming.

Working with a single microphone input for outgoing speech and—uniquely—also processing incoming signals from the other parties in a connection, ClearVox ENC effectively separates speech from both continuous and transient background noise. Versions are available for Ceva-BX1, BX2, and SensPro2 DSP cores and for ARM MCUs.

Key features

  • Embedded, neural-network-based algorithm requires only a single microphone.
  • Local processing means no Internet latency or cloud-AI privacy issues.
  • Bidirectional processing cancels noise on both outgoing and incoming signals.
  • 16 kHz sampling rate and 16-bit resolution ensure high speech quality and cancellation of broad-spectrum and high-frequency environmental noise.
  • Optimized code requires little memory space or processing power, allowing use in small systems even in conjunction with other applications.

Block Diagram

Ceva-ClearVox ENC block diagram

Benefits

  • Using a trained, optimized, and highly compact neural network, ClearVox ENC effectively separates clear voice signals from both continuous and transient ambient noise, yielding intelligible speech even with adverse environmental noise. The ClearVox ENC algorithm requires only a single local microphone, and uniquely processes both outgoing and incoming speech. Thus both parties in a conference or gaming environment benefit from the ENC.
  • The software requires minimal memory space and computing power, so it can be run effectively even in quite small systems, and in conjunction with other applications, such as Ceva-RealSpace.

Applications

  • Consumer IoT
  • Automotive
  • Mobile
  • PC 

Video

MWC BARCELONA 2018 CEVA CLEARVOX DEMO
Far Field Voice Controlled smart speaker reference design

Files

Product Sheet Download

Note: some files may require an NDA depending on provider policy.

Specifications

Identity

Part Number
Ceva-ClearVox ENC
Vendor
Ceva, Inc.

Provider

Ceva, Inc.
HQ: USA
The Smart Edge runs on Ceva! Ceva is the leader in innovative silicon and software IP solutions that enable smart edge products to connect, sense, and infer data more reliably and efficiently. At Ceva, we are passionate about the smart edge. Providing the technology and market expertise our customers need to be successful is what we do best, and we’ve been doing it for over 30 years. With the industry’s only portfolio of comprehensive communications and scalable edge AI IP, Ceva powers the connectivity, sensing, and inference in today’s most advanced smart edge products across consumer IoT, mobile, automotive, infrastructure, industrial, and personal computing. More than 17 billion of the world’s most innovative smart edge products from smartphones to drones to cellular base stations and more are powered by Ceva. We create innovative technologies that help our customers turn great ideas into extraordinary products. We license our portfolio of wireless communications and scalable edge AI IP to our customers, breaking down barriers to entry and enabling them to bring new cutting-edge products to market faster, more reliably, efficiently, and economically. Ceva is a trusted partner to over 400 of the leading semiconductor and OEM companies including Actions, Artosyn, ASR, Atmosic, Autotalks, Beken, Bestechnic, Brite, Broadcom, Celeno, Ceragon, Cirrus Logic, Dialog Semiconductor, DSP Group, Espressif, FujiFilm, GCT Semi, iCatch, InPlay, Intel, Itron, Leadcore, LG Electronics, Mediatek, Microchip, Nextchip, Nokia, Novatek, NXP, ON Semiconductor, Optek, Oticon, Panasonic, RDA, Renesas, Rockchip, Rohm, Samsung, Sanechips, Sharp, Siflower, SigmaStar, Socionext, Sony, Socionext, Sonova, STMicroelectronics, Toshiba, Unisoc, Vatics, Yamaha and ZTE all leverage Ceva’s industry-leading IP. These companies incorporate our IP into application-specific integrated circuits (“ASICs”) and application-specific standard products (“ASSPs”) that they manufacture, market and sell to consumer electronics companies. Headquartered in Rockville, Maryland, Ceva has over 400 employees worldwide, with design centers in Israel, Ireland, France, United Kingdom, United States, Serbia, and sales and support offices located in Europe, the U.S. and throughout Asia. Ceva is a sustainable and environmentally conscious company, adhering to our Code of Business Conduct and Ethics. As such, we emphasize and focus on environmental preservation, recycling, the welfare of our employees and privacy – which we promote on a corporate level. At Ceva, we are committed to social responsibility, values of preservation and consciousness towards these purposes.
Contact Ceva, Inc.

Learn more about Edge AI Accelerator IP core

RISC-V Based TinyML Accelerator for Depthwise Separable Convolutions in Edge AI

While lightweight architectures like MobileNetV2 employ Depthwise Separable Convolutions (DSC) to reduce computational complexity, their multi-stage design introduces a critical performance bottleneck inherent to layer-by-layer execution: the high energy and latency cost of transferring intermediate feature maps to either large on-chip buffers or off-chip DRAM. To address this memory wall, this paper introduces a novel hardware accelerator architecture that utilizes a fused pixel-wise dataflow.

Accelerating Your Development: Simplify SoC I/O with a Single Multi-Protocol SerDes IP

Enter the Multi-Protocol SerDes (Serializer/Deserializer)—a flexible, reusable IP block that allows a single PHY to support multiple serial communication protocols, such as PCIe, SATA, Ethernet, USB, and more. This approach enables SoC vendors to meet diverse customer requirements and application needs without redesigning I/O for each target market.

Frequently asked questions about Edge AI Accelerator IP cores

What is Neural-network-based noise cancellation?

Neural-network-based noise cancellation is a Edge AI Accelerator IP core from Ceva, Inc. listed on Semi IP Hub.

How should engineers evaluate this Edge AI Accelerator?

Engineers should review the overview, key features, supported foundries and nodes, maturity, deliverables, and provider information before shortlisting this Edge AI Accelerator IP.

Can this semiconductor IP be compared with similar products?

Yes. Buyers can compare this product with similar semiconductor IP cores or IP families based on category, provider, process options, and structured technical specifications.

×
Semiconductor IP