Cyberon STT uses a deep learning algorithm to train the model. It also supports 16KHz above and 8KHz voice data (phone), which can provide split-second voice recognition on a range of devices, as well as recognition of recording files.
It has the following characteristics:


  • Flexible deployment options: Provide cloud-based speech recognition or be installed on the enterprise server.
  • Customized language model: Allows customized vocabulary for different application areas.
  • Support multiple front-end devices : Release SDK and API which can support App, Web, Smart Speaker and various IoT front-end devices.