Cloud und Künstliche Intelligenz: Virtuelle Persönliche Assistenz mit ODP S3

Die ontologiebasierte Dialogplattform ODP S3 von SemVox integriert KI-Technologien auf verschiedenen Ebenen. Basierend auf einer ontologischen Ebene führt ODP S3 verschiedene (auch cloudbasierte) Dienste zusammen, die es dem Assistenzsystem über Machine-Learning- und Reasoning-Verfahren ermöglichen, logische Schlüsse zu ziehen und die so ein intelligentes Systemverhalten hervorrufen.

Die cloudbasierte NLU (Natural Language Understanding) ermöglicht es ODP-Systemen, natürlichsprachlich zu verstehen und zu reagieren.

Das automatische logische Schließen (Reasoning) formalisiert menschliches Wissen und transferiert es in eine maschinenlesbare Form. Machine-Learning-Methoden und Cognitive Computing befähigen ODP S3, aus dem Kontext und dem Nutzerverhalten zu lernen, um die Systemreaktion anzupassen und auf den Nutzer zugeschnitten zu aktivieren (Profiling).

Mehr Infos unter Proaktive Assistenz

This video shows our SemVox team member Thomas acting as an assembly trainee or worker in an interaction with an instructor robot.

The project has been carried out as a joint research between the htw Saar and the University of Calgary.

In industrial working environments, especially in the assembly industry, it is critical to learn how to carry out a new task and reproduce it within the context of an assembly manufacturing line. Before workers can be employed in a production line, they have to become familiar with all steps and aspects of the assembly process. For example, for a gearbox, manufactured in a production line, workers have to train this complex process several times in order to ensure a flaw- and frictionless assembly.

The proposed approach integrates humanoid robots into a learning system, wherein a robot assumes the task of an overseer during an industrial learning process. By establishing human-robot collaboration for learning processes, workers will be able to train an assembly process in a face-to-face setting with the robot.


Assembly and Voice Dialog Demo with Baxter(TM) Robot

SemVox Interactive Manual – Embedded Installation on Google Glass

The SemVox technology is a wholly new option to provide relevant information to production and manufacturing processes. Thanks to this innovative platform, such a system gives users accurate access to information at any stage of the production or maintenance process. Voice control facilitates access to information by means of hands-free, intuitive and efficient interaction. Unlike with conventional handbooks, the interactive access to production knowledge also facilitates the flexible integration of videos, interactive image sequences, spoken language and various other media.

With ODP S3, SemVox covers the entire spectrum of electrical automation of production and logistics. Integrators and suppliers of voice control and monitoring solutions will find SemVox ODP S3 to be the perfect platform for all their customer projects.

Using the SemVox ODP S3 technology, voice control solutions and multimodal proactive assistance systems can be developed for a variety of use cases and application scenarios in industry, construction and manufacturing, such as production processes or maintenance procedures in production lines.

Industrie und Logistik

The ODP S3 demonstration system exemplifies how in-car devices can be accessed by speech via the internal CAN-bus (controller-area network bus) of the car. The technology behind the ODP4Car demonstration system is a multimodal dialogue system based on the ODP (Ontology-based Dialogue Platform) framework by SemVox GmbH. The ODP framework permits a speech-based access to functions like controlling the air condition, the audio system or the windows via speech. This goes beyond usual ways of controlling individual functions as the system is able to understand and process even such general utterances like "I'm cold" and then to trigger the appropriate actions, in this case an increase of temperature in the car.
For more information see Automotive.

ODP S3 – Einfach mit dem Auto sprechen