Fraunhofer Institute for Digital Media Technology IDMTWhether machines, media or household appliances, building technology or the telephone – are you looking for a solution that allows your customers to operate it naturally? Ambient voice control from the Fraunhofer IDMT's Oldenburg Branch for Hearing, Speech and Audio Technology HSA enables simple, contactless and reliable operation of complex technical devices or systems on demand. The software solution can be integrated directly into your application and is adaptable to a wide range of application scenarios and end devices.
Speech is the most natural form of communication
Efficient and safe collaboration between humans and machines through robust voice control
© Fraunhofer IDMT/Anika Bödecker
Voice Control
Benefits for your customers
Intuitive voice commands for Industry 5.0
The Fraunhofer IDMT in Oldenburg develops application-specific, robust and intuitive voice control solutions for use in production. Voice control is easy to integrate and also works without an internet connection. The voice control solutions can be integrated into the machine controller as well as Windows or Linux platforms. It reliably recognizes voice commands even under the demanding acoustic conditions of industrial production. This results in lower production costs and reduced walking distances when operating multiple machines.
Optimal support in everyday life thanks to reliable operation
The voice control of the Fraunhofer IDMT enables users who are not familiar with technology to benefit from complex technical systems in their everyday lives. This is important, for example, in assistance systems for people who need both hands for other activities or elderly or physically impaired people. Especially for these users, systems must be easy to operate and function reliably.
Voice input for media applications
Voice control makes classic operating elements or a remote control superfluous. Instead of per device, smart home solutions can also be controlled centrally by voice input, e.g. via a central media control system with corresponding interfaces from our partners.
Benefits for your company
Customised solution
Our voice control systems are scalable to any degree and can be individually adapted to your application scenario. One focus of Fraunhofer voice control is on robust systems with ambient microphones. The microphone does not have to be immediately near to the mouth. This allows a customised connection to your applications, end devices or even existing systems. The vocabulary is freely definable.
Opening up new customer groups
Fraunhofer IDMT's voice control breaks down boundaries. The user communicates naturally by speech and receives immediate feedback. This makes complex systems and devices reliably operable even for target groups that are not familiar with technology, or they can be controlled without contact in sterile environments.
Cost savings thanks to efficient development
Benefit from the expertise of the Fraunhofer IDMT's Oldenburg Branch for Hearing, Speech and Audio Technology HSA and integrate the latest speech recognition technologies into your products, even without your own experience in the field of speech processing.
Reliable operation for satisfied customers
A new, scalable detection method for voice commands is characterised by reliability and robustness against noise and room influences, even with microphones installed in the room (ambient) and at a distance from the user. It reduces the false alarm rate, expands the application possibilities and increases customer satisfaction.
Maximum stability and data protection
Our speech recognition systems run directly on the devices to be controlled or as a service within the company's own infrastructure. The services are therefore independent of external connections or services, highly available and data processing can take place entirely in-house. This means that even the highest data protection requirements can be met.
Specification
- Scalable range of functions
- Connection to various applications and end devices possible
- Runs on different platforms (PC, embedded systems, Cloud/Docker)
- Dialogue-capable
- Speaker-independent use
- Ambience and individual microphoning
- Multi-channel miking
- Robust against external noise and room influences such as traffic, grass mowers, room reverberation and echo
- Multi-channel signal processing
- Directional filtering (beamforming)
- Scalable vocabulary
Technical Data
- Scalable, modular combination of the various signal enhancement technologies (directional filtering, de-noising, echo reduction, reverberation suppression, etc.) with the speech control technologies.
- Use of innovative, auditory-motivated speech features from research for speech recognition to maximise robustness against acoustic disturbances
Marvin Norda about controlling machines by voice commands
Privacy warning
With the click on the play button an external video from www.youtube.com is loaded and started. Your data is possible transferred and stored to third party. Do not start the video if you disagree. Find more about the youtube privacy statement under the following link: https://policies.google.com/privacyMarvin Norda, coordinator of the industrial working group "Audio Technology for Intelligent Production AiP", talks about our speech recognition solutions for industrial processes.
Demo: Talking to machines
Privacy warning
With the click on the play button an external video from www.youtube.com is loaded and started. Your data is possible transferred and stored to third party. Do not start the video if you disagree. Find more about the youtube privacy statement under the following link: https://policies.google.com/privacyWe develop application-specific, robust and intuitive speech recognition solutions for use in industry and production. Speech recognition can be easily integrated and also works without an internet connection. Voice commands can be reliably recognized even under demanding acoustic conditions.