Beamforming with Microphone Arrays for Optimal Localisation

vicMSES - Multichannel Speech Enhancement Suite.

Beamforming is used, if single channel noise suppression methods have their limitations or if a spatial separation of speech and background noise is necessary. It is based on the spatial distribution of several microphones (arrays). Static or adaptive algorithms (beamforming, beemsteering) enable the electronic post processing of the microphone signals. We summarised them in a library called Multichannel Speech Enhancement Suite (vicMSES).

PURPOSES

  • Separation of simultaneous signals (e. g. noise and speech, speech signals of several speakers)
  • Elimination of disturbing room conditions (e. g. reverberation)
  • Acoustic localisation of sound sources

PRODUCT FEATURES

  • Complete library of algorithms for beamforming, dereverberation and source separation
  • Platform-independent, proven code with high portability for embedded platforms (ADSP Blackfin, SHARC; TI C6000; i.Mx6; ARM)
  • Simulation and measurement tools for the design of customer-specific microphone arrays
  • Graphic parameterisation and evaluation tools

Applications

  • Hands-free systems for cars and commercial vehicles
  • On-board communication (tram, bus, train, ship, airplane)
  • Communication between control stations and emergency vehicles
  • Acoustic room monitoring
  • Conference technology
  • Building communication
  • Telecommunication devices, Telekommunikationsgeräte, smartphones and tablets