• Voice Recognition

e.sigma’s design strategy ensures that its Simulators and Systems are intuitive, user-friendly and from a user perspective as close to reality as possible. For this reason it became necessary that existing “human- machine-interface” (HMI) modules, consisting of manual keyboard and optical feedback input options, be enhanced through addition of voice interaction (multimodal) modules. For this purpose e.sigma developed a proprietary, configurable voice-recognition/voice-response communications module through which Air-traffic controller-students and their Instructors can voice-activate simulation functions to direct virtual Aircraft.

  • Core components of the e.sigma-Voice recognition / voice control modules:
  • Autonomous language recognition module
  • Status dependent syntax
  • Speaker voice adaptation
  • Emotional-Text-to-Speech
  • Dialog management system
  • Configuration-module


The language-independent recognition- module is first programmed to the exact requirements of the customer and then trained with the client’s application related audio data. This domain specific training method guaranties a strong voice-recognition result. In addition audio-files of the events are recorded, during system activity, enhancing further the systems acoustic module database.

The prerequisite for good voice-control results is first-rate voice-recognition. This is achieved through application of a status dependent syntax protocol. Background information from diverse Simulations, activate the required, situation specific syntax and phrase. Using this method one can minimize the identification complexity and optimize its voice-recognition capacity. The ruggedness of voice-controlled simulation-systems is further enhanced by an unmonitored voice-adaptation-process.  Applying this e.sigma process improves the recognition capacity of the individual speaker significantly.

Emotional-Text-To-Speech (TTS) is another important system-module. It allows common phrases such as ICAO, to be sent to the student, along with configured, subject related answers and emotional content.

The core component of the multi-modular system-control is the Dialog-Management-System. Its function is to receive and digest the identified hypotheses of the speaker, then responding by issuing the order to select the appropriate simulation-function. The possibilities achieved through its proprietary configurable multi-modular interface puts e.sigma in a unique position. The user is given the freedom to define orders using voice or text commands and the option to expand and modify the syntax-pool.

  • Rapid, simple implementation of voice-recognition modules
  • Enables easy learning of commonly used domain related phrases, through practical application of relevant linguistic rules in the simulation.
  • Contains all prevalent Dialogue used in civil and military aviation