The technology behind our solutions

Our innovative technologies are the basis for our products and solutions. In order to develop these technologies, we have brought together state-of-the-art concepts from the area of artificial intelligence with the latest findings from the area of speech technology. We have rendered it into a market-ready base technology with which we can develop future-oriented intuitive applications in the area of human-computer interaction for you.

Dialog processing - With ODP, SemVox markets a modular platform for the realization of innovative user interfaces. ODP, which relies a hundred per cent on Java and XML, is web-based and can easily be integrated into existing environments, because it supports widely accepted standards such as MRCP/SIP, EMMA and SSML. With ODP, widespread GUI-frameworks such as Ajax, Flash/Flex or JavaFX are as easily integrated as new types of modalities (such as multi-touch, gesture recognition or virtual characters). Client libraries for the mobile platforms Android, iPhone and Windows Mobile are available as well – they work with the web-protocols HTTP and RTP.

More relevant information on the ODP platform can be found here. 

Semantic Technologies - The processing within the single components of the ODP-framework is realised on a fully integrated semantic level that abstracts from technical details of the underlying contents and services. Ontologies are the basis for the semantic modeling in our technology which, in turn, forms the basis of this processing. Ontologies are formal representations of a number of concepts and their respective relations to each other within the frame of a certain application domain. We use them to model the knowledge a systems needs for a certain application or application domain. The following example helps to illustrate the issue:

Multimodality - A key aspect of the ODP-technology is the robust and fast processing of multimodal input and output. Multimodal means, that a user has different means of controlling a system: via speech, touch, multi-touch, mouse, keyboard etc. Multimodal processing within SemVox products even allows true combination of several different input modalities. For example, a user can point to an element on a touch-screen with the cursor and accompany the gesture with the words: “Give me more information on that”. ODP is able to interpret several user inputs from different channels simultaneously and derive the user´s meaning from them.

This provides intuitive access to complex services and contents – depending on the situation and the user´s personal preferences, he can choose between different input modalities. Especially the use of natural language as input modality offers a target-oriented access to available functionalities, without having to deal with complex menu structures.

Multilinguality - Interactive speech-processing systems such as speech recognizers or speech synthesizers have reached a high-quality standard in recent years. In a world that is becoming more international by the day, the demands on a speech-processing system are high. It is necessary for such a system to be able to react to multilingual input in an adequate way. Especially when dealing with media like videos or music, the processing of multilingual data such as movie or song titles or the names of artists is a challenge for speech-centered interaction systems. SemVox has developed a technology that makes it possible to create appropriate enhancements for the speech recognizer and speech synthesizer components for any domain that involves multilingual data.