Spoken dialog systems for business and domain knowledge

We develop speech-driven dialog systems with a special focus on domain-specific knowledge for application in various fields of business and industry. Combining state-of-the-art components for Speech Recognition, Question/Answering via Knowledge Graphs and Speech Synthesis, our technologies address in particular the concrete challenges and needs of enterprises and B2B applications.

Dialog Systems

Intuitive interaction in B2B applications

Speech assistants are used in more and more areas of life, enabling intuitive interaction with technology, providing service and information. They are not only useful in everyday life, but also offer companies great potential for facilitating human-machine interaction and offering completely new services to customers.

We develop speech-driven dialog systems with a special focus on domain-specific knowledge for application in various fields of business and industry. Combining state-of-the-art components for Speech Recognition, Question/Answering via Knowledge Graphs and Speech Synthesis, our technologies address in particular the concrete challenges and needs of enterprises and B2B applications.

Technologies “made in Germany” ensure technological sovereignity

Moreover, these technologies ensure technological sovereignty, data can be stored and processed within secure data spaces. “Informed Machine Learning” methods, developed within the Fraunhofer Cluster of Excellence Cognitive Internet Technologies CCIT (Center for Machine Learning), make sure that the systems can even be trained on small data sets.

 

Our offer

  • Our team develops dialog systems for domain-specific knowledge for applications in various fields of business and industry.
  • We offer a set of domain-adaptable state-of-the-art components for spoken dialog systems, ranging from speech recognition to dialog management and speech synthesis.
  • We customize the systems for your specific requirements and help you leverage semantically structured knowledge from different data sources.
  • In addition, we support you in the development of new business models that involve dialog systems.
  • We ensure digital sovereignity, data can be stored and processed in secure data spaces.

State-of-the-art components for B2B dialog systems

© Fraunhofer IAIS

1. Speech Recognition – Speech to Text

The task of speech recognition is to convert spoken language into text. The linguistic modelling is realized with the help of machine learning algorithms: The language model is trained in a certain domain of knowledge so that the pronunciation of the words is technically correct. In addition, the system learns frequent word sequences and dialects.
 

2. Question Answering via Knowledge Graphs

For specific application areas, it is important to integrate domain-specific knowledge into the systems. For this, we use Question/Answering technologies (QA). The challenge on the technological side is to understand the requests of users and to answer them with the help of domain-specific information from different sources.

The system has to recognize which knowledge domain – for example logistics, emergency management or business processes – the query is about. Technical knowledge is complex and can best be organized in a knowledge graph – an intelligent network which stores data and information and makes them machine-readable.

In this way, an intent classifier recognizes the topic and searches for the suitable answer. With the help of verbalization techniques, the system then ensures that the answer is formulated. Tailored to the respective knowledge domains, our experts train machine learning algorithms using example dialogs and question-answer pairs.
 

3. Speech Synthesis – Text to Speech

Speech synthesis completes the dialog system by converting the textual information from the QA into spoken language. This process is also  based on machine learning: With the help of transcribed audio recordings, the model learns to generate a waveform from letters. In this way, the users hear the answers from the loudspeaker.

 

Use cases for spoken dialog systems

© Fraunhofer IAIS

© Fraunhofer IAIS

Our dialog systems focus on domain-specific knowledge and can be trained for various fields of application. The technologies can be adapted to many application areas and are already prototypically or even completely integrated in various contexts:

  • The technology can support business processes or help to quickly find information in corporate or media archives.
  • Integrated into the car, the dialog system can serve as an interactive city guide, answering questions about certain points of interest.
  • Medical devices, equipped with a speech assistant, can be controlled by voice input. In combination with methods of knowledge extraction from medical documents, experts can receive crucial support in making diagnoses.
  • Dialog technology can save valuable time, for example when emergency teams use headsets to get quick answers to life-saving questions.