Voice activation, far-field voice pickup, voice communication, audio processing, and sound sensing

Significant advances in Automatic Speech Recognition (ASR) have led to an abundance of devices and applications that use speech as their main user interface. Microphones are no longer regarded as mere voice input devices for phone calls but rather as acoustic sensors for human-machine interaction and for acoustic sensing of the environment. We have come to expect high-quality ASR as well as other audio-related functionalities, such as noise-free voice calls and immersive high-definition audio, from our devices.

As Artificial Intelligence brings us closer to creating human-like machines, we must develop similar audio-processing abilities to the ones we have as human beings.
Application specific audio processors and specialized software packages are the only viable way to accomplish this.

  • Voice User Interface is a key experience in almost any consumer device today.
  • Dynamic and adaptive sound sensing enables machines to act upon recognizable audio signals.
  • High quality speech recognition and voice communication even in noisy environments.
  • 3D audio processing technology delivers a panoramic audio experience and realistic sense of space.

The ability to capture, process, and reproduce audio signals is fundamental to mobile devices, headsets, wireless speakers, smart home devices, automotive infotainment systems, and voice enabled IoT. To enable these devices a scalable audio processor DSP is required.
Today’s range of voice enabled devices require ultra-low-power, high-performance processing for speech recognition and reproduction, advanced multi-microphone voice capabilities, and always-listening functionality. Because they employ multi-microphone arrays for far-field voice pickup, smart speakers and other smart mobile devices must be optimized for the best combination of performance and power consumption. Finally, premium audio experience for home entertainment and automotive infotainment systems demand the best possible audio fidelity.

The key to meeting these requirements lies in embedding dedicated audio DSP cores. The use of flexible audio and voice DSP IP architectures and optimized software packages for it, provides the following:

  • Highest performance DSP architecture, delivered with small die size, high code density, and high processing capability
  • Enable SoC designers to select the optimal implementation in terms of silicon area, power consumption, and operating frequency
  • Proprietary CEVA designed audio/voice/speech software solutions that are optimized for the DSP reduce time to market and cost of full system
  • Ecosystem based software modules that are available off-the-shelf, and development platforms to speed application development and prototyping

 

  Audio

  Codecs

  Voice

  Codecs

  Dolby

  HD-Audio

  DTS

  HD-Audio

LC3

MP3

MP3Pro

Ogg Vorbis

FLAC

AAC LC

HE AAC V1

HE AAC V2

HE-AAC V2 5.1

MPEG4 BSAC

WMA

SBC

CELT

DRA

EVS

AMR-WB

AMR-NB

HR

FR

EFR

EVRC

QCELP

SILK

OPUS

iLBC

mSBC

AMBE

G.7xx

Dolby Atmos

Dolby AC4

Dolby TrueHD

Dolby Digital Plus

Dolby Digital decoder       (AC3)

Dolby Digital encoder   (DDCE)

Dolby MS11

Dolby MS12

Dolby Volume

Master Audio

High Resolution

Low Bit Rate

Extended Surround   (ES)

DTS 96/24

DTS Digital Surround

DTS Transcoder

DTS M6

DTS M8

 

Check out the related product links below to find the perfect audio DSP IP solution for your needs.

Target Applications

Voice Assistant

Voice as a Human Machine Interface, starting with voice trigger, voice commands and up to natural language understanding at the edge, requires ultra-low-power and always-on solutions, using far-field voice pickup, beamforming, and acoustic echo cancellation, to allow intelligible voice assistant conversations.

Headset

From corporate communication VoIP headsets to in-ear truly wireless earbuds, efficiently tailored noise reduction, acoustic echo cancelation and active noise control and 3D Audio, are essential for enhanced user experience and seamless connection to voice assistants.

Mobile

Efficient, "on the move" sound processing is critical for voice communication in smartphones and tablets. Enabling a clean and noise-free voice signals involves the use of complex codecs such as EVS, as well as advanced noise-reduction techniques.

Wearable

Smartwatches, smart glasses and other wearable devices are very power-constrained, which calls for ultra-low-power processing to facilitate voice control and enhanced audio experience.

Smart Home

As home devices get ever smarter, smart speakers, security cameras, Digital TVs, set-top boxes, smoke alarms, and many other home appliances introduce far-field voice control, sound sensing, and high-quality audio.

Automotive

Automotive infotainment systems integrate voice processing and speech recognition alongside high fidelity audio playback in a challenging environment with a high noise floor.

Our 2nd generation Cupola360 SoC is our first product to incorporate real-time multi-image stitching video and powerful audio processing capabilities, making it perfect for video conferencing applications. Through the comprehensive cooperation between CEVA and ASPEED, our customers can enjoy the most superior video & audio functionality and we are glad to have CEVA as our strong and trusted partner.
Chris Lin, Chairman and President of ASPEED Technology

CEVA Voice-control enabled solutions for IoT

End-to-end voice control solutions for IoT based on CEVA’s ClearVox multi-mic noise reduction and WhisPro voice user interface speech recognition SDKs. Available for CEVA-BX and SensPro Audio/AI DSPs and Arm CPUs.