Technical transmission of speech is often subject to superposed interference noise and reverberation, e.g. in the case of railway station announcements or when using mobile phones. The Project Group for Hearing, Speech and Audio Technology offers real-time signal processing methods that analyse and represent speech intelligibility under conditions with unknown and variable interference noise. For this purpose, the current acoustic situation is detected using a microphone. An adaptive signal processing algorithm developed at the Fraunhofer IDMT is then able to automatically adapt the speech signal to the background noise. By taking into account the latest hearing research findings, this makes it possible to achieve a high level of hearing comfort even for signals with limited bandwidth and for people with impaired hearing.
© Photo iStock
Real-time monitoring of speech intelligibility