Neural network-based speech recognition technology for voice assistants and IoT devices

WhisPro™ is a neural network based speech recognition software package, allowing customers to add voice trigger and commands to voice-enabled IoT devices. WhisPro is targeting the rapidly growing use of voice as a primary human interface for intelligent cloud-based services and edge devices.

WhisPro provides an always-listening multi-trigger phrase capability, allowing users of headsets, smart TVs, smart speakers, smartphones and other voice-enabled devices to interact with cloud-based voice assistant services as well as allowing voice activation of edge devices without cloud connectivity.


WhisPro is optimized for always listening devices, enabled by voice trigger and commands

High recognition rate of more than 95% without the need for cloud connectivity
Scalable RNN speech AI technology supports simultaneous multi trigger phrases
Built-in noise immunity for improved user experience in noisy environments

Main Features

  • Customized voice trigger and commands service to enable voice control for a range of use cases and end markets
  • Speaker independent, pronunciation agnostic, and support for multiple languages
  • Integrated with Google’s TensorFlow Lite for Microcontrollers, accelerating the development of small footprint voice assistants and other voice controlled IoT devices
  • Operates in tandem with CEVA’s ClearVox front-end voice processing software technology enables a complete speech recognition solution with state of the art noise reduction and AEC for barge in capabilities
  • The WhisPro SDKit provides guidelines and tools to integrate and test the WhisPro technology in a target system
  • WhisPro is available for CEVA’s DSPs: CEVA-BX1, CEVA-BX2 and SensPro family, as well as Arm CPUs

Block Diagram