Fraunhofer IDMT with telecommunications solutions for speech intelligibility and recognition at CCW 2019

Recent / 13.2.2019

From 19 - 21 February 2019 Fraunhofer IDMT in Oldenburg will present communication solutions for better speech intelligibility and automatic speech recognition in telephony at Call Center World (CCW) in Berlin (Stand C11, Hall 2) with real added value for customers and staff.

© Thinkstock/Wavebreakmedia Ltd

Fraunhofer IDMT with telecommunications solutions for speech intelligibility and recognition at CCW 2019.

Oldenburg, 13.02.2019. Call centre staff are advisors, troubleshooters and agony aunts all at once. This demands a high level of attention in order to guarantee the best possible customer experience. Acoustic conditions vary from call to call, depending on the connection, background noise, volume, ambient reverberation or other factors.


Better sound, less noise!

So that call centre agents can set the sound configurations for telephone conversations to meet their individual needs, experts at Fraunhofer IDMT in Oldenburg have developed a convenient application based on the interaction of a wide variety of algorithms. Whether analogue or digital telephony is used makes no difference, the device is easily installed between the telephone system and the agent’s headset or receiver. Individual presets make it possible to adjust the device to personal hearing preferences and habits. Depending on the preset, incoming audio signals are corrected in such a way that the desired sound and preferred volume are achieved.


Dr. Jan Rennies-Hochmuth, Head of Group, Personalized Hearing Systems, at Fraunhofer IDMT in Oldenburg, explains: “Suppressing background noise or other disturbing factors in telephone calls noticeably relieves the strain on the staff engaged in such calls. Since hearing perception is always also a subjective sensation, it was particularly important in our development work to make individual settings possible.”


At CCW 2019, Fraunhofer IDMT in Oldenburg will present a demonstration model of its system for the optimization of speech intelligibility in telephony. Visitors will listen to different sound scenarios, with the aim of subjecting the device and the improved signal to a live test by switching it on or off.


Talk to me! Automatic voice assistants in customer dialogue

Apart from tackling challenging acoustics, computer-based processes can also reduce recurring or administrative tasks for staff at telephone workstations. Examples are the analysis of call contents through automatic speech recognition or the requesting of basic information, such as name, customer number and date of birth, prior to the actual person-to-person telephone call itself.

Here too, Fraunhofer IDMT in Oldenburg is a source of help with its developments in the area of “Automatic Speech Recognition”. Through the textualization of speech (speech-to-text), algorithms can analyze call contents and recognize pre-defined keywords so that these can be attributed to specific topics. This makes it possible to answer frequently asked questions automatically or even conduct complex dialogues, making work especially in call and contact centres much more efficient.


“Intelligent voice input and dialogue systems are an important development for man-machine interaction. The use of computer-assisted telephone dialogue in customer contact is a good example of how this technology can deliver genuine added value in the working world. In this context, we’re aiming to create an interlocutor who appears as natural as possible with a very low misrecognition ratio,” says Dr. Stefan Goetze, Head of Group, Automatic Speech Recognition at Fraunhofer IDMT.


At CCW 2019, Fraunhofer IDMT will demonstrate by way of example and with the help of a chatbot system how IDMT’s speech recognizer can be used in personnel recruitment. In a spoken dialogue with the system, visitors can take part in a fictitious job interview.


Discover pioneering developments by the Hearing, Speech and Audio Technology Division (HSA) at the Fraunhofer Institute for Digital Media Technology IDMT in Oldenburg at CCW from 19 - 21 February 2019, Hall 2, Stand C11. In the area of “Automatic Speech Recognition”, Fraunhofer IDMT will present by way of example a chatbot demonstration model for personnel recruitment. In addition, in the area of “Personalized Hearing Systems” visitors can test an individually adjustable system for better speech intelligibility in telephony. They can listen to several sound scenarios and switch the system on or off as required.


Hearing, speech and audio technology at the Fraunhofer Institute for Digital Media Technology IDMT in Oldenburg

The objective of the Hearing, Speech and Audio Technology Division (HSA) at Fraunhofer IDMT in Oldenburg is to transpose scientific findings in the field of hearing perception into technological applications. Its applied research priorities are the improvement of sound and speech intelligibility, personalized audio reproduction and acoustic speech and event detection. Application fields include consumer electronics, transport, the automotive sector, industrial production, safety, telecommunications and healthcare. Through scientific partnerships, Fraunhofer IDMT-HSA has close links with Carl von Ossietzky University, Jade University and other institutions engaged in hearing research in Oldenburg. Fraunhofer IDMT-HSA is a partner in the »Hearing4all« Cluster of Excellence.