Will machines talk with us?
Language is the natural way how people communicate. Why not let the machine talk to the operators, too?
With advances in machine learning and speech recognition, we can take advantage of these technologies and connect machines with a human operator by natural language communication. Machines and computers process huge amounts of data that humans can no longer handle. In addition to visual interfaces, voice communication makes it possible to extend the interaction of the operator with the machine and share important, extracted data with the operator.
Beyond, SABOT will integrate also other areas of machine learning, e.g. image processing, the learning of operators’ behaviour and its prediction.
Main SABOT advantages
- Machines are becoming increasingly complex, but they should be easy to operate. The quantity of different questions and commands which are addressed over touch controls like button or screens can be reduced to single instruction over the voice.
- Voice communication is possible hands-free at a distance and as well in the dark, is therefore fully usable even in adverse working environments.
- Operators can observe the process while giving instructions and asking questions, without eye contact with the operator terminal.
- Voice interaction is faster than touch screen inputs can be.
- Aspects of leadership, the experience gained from senior operators can be used to guide younger operators.
- And last but not least: because we will get used to speech input as a standard very quickly.
SABOT empowers all kinds of machines to speak
SABOT serves as a VUI (Voice User Interface) for machines of all kinds, can be retrofitted and is developed for commercial use where data security, operating safety and use in industrial environments are required.
This is how much AI is in SABOT
- SABOT's voice user interface (VUI) supports common STT (Speech-to-Text) and TTS (Text-to-Speech) services.
- Intentions (queries and control commands) can be expressed in common speech and are understood, thanks to Natural Language Understanding (NLU).
- SABOT communicates with operators in all common languages.
- SABOT leverages the knowledge base of large transformers.
- Noise cancelling by means of AI suppresses disturbing ambient noise.
- Different user profiles have been implemented - experienced and less experienced machine operators.
- A sentiment analysis automatically detects whether machine operators have a positive or negative attitude and adapts the communication accordingly. This ensures a high level of user acceptance.
- User-defined scenarios for specific machines and industries are trained and optimized in operation with Conversation Driven Development (CDD).
- Voice recognition not only helps with authentication, it also allows operator guidance tailored to the person ("I want a coffee in ten minutes!" - "Same as always, Christian?").