VOCAL Print Logo
Beamforming >  Acoustic

Acoustic Beamforming

Acoustic beamforming can efficiently enhance the speech of interest while suppressing interference, background noise, and reverberation. It allows people to move freely around without wearing or holding a microphone. Acoustic beamforming provides the option to capture the individual speech from each talker and generate separate voice streams, i.e. it enhances the signal flow from the talk of interest to specific individuals and allows background noise (other talk, motors, movement, etc.) to take place.

VOCAL's software implementation of acoustic beamforming allows for uniform linear array, uniform circular array, and uniform square array orientations. The maximum number of microphones that can be processed by the acoustic beamforming software for real-time applications would be limited by the capabilities of the processor. The software package contains a source localization algorithm to estimate the Direction of Arrival (DoA) of the desired signal and to dynamically steer the beamforming algorithm in the appropriate direction to capture this signal.

VOCAL's acoustic beamforming includes, but is not limited to, the following algorithms.

  • Delay-and sum beamforming
  • Linearly-Constrained Minimum-Variance (LCMV) beamforming
  • Time-Domain Generalized Sidelobe Canceller (TD-GSC)
  • Frequency-Domain Generalized Sidelobe Canceller (FD-GSC)
  • Robust Generalized Sidelobe Canceller (R-GSC)
  • Degenerate Unmixing Estimation Technique (DUET) algorithm

These algorithms are applicable to a wide range of the array configurations that exist in reverberation environments and can compensate for errors in the Direction of Arrival (DoA) estimation of the desired signal. The selection of the algorithm is based on the requirements of the application and the available hardware configuration (e.g. the number of microphones, the capabilities of the desired processor). The diagram of the most commonly implemented acoustic beamforming, Time-Domain Generalized Sidelobe Canceller (TD-GSC), is shown in Figure 1. The quality of the speech can be enhanced by integrating any number of optional features with the beamforming scheme, based on the requirements of the application and the acoustic environments.

Time-Domain Generalized Sidelobe Canceller (TD-GSC)

Figure 1: Time-Domain Generalized Sidelobe Canceller (TD-GSC)

  Standard Features
  Scalable number of microphones
  Scalable sampling rate
  Scalable frame size
  Supports both near field and far field
  Supports uniform linear array, uniform circular array, and uniform square array
  Fixed-angle or dynamic steering
  Multi-channel operation
  Fully adaptive system adapts to changing acoustic environment

  Optional Features
  Full-duplex Acoustic Echo Cancellation (AEC)
  Stationary and Non-stationary noise suppression
  Active noise controller
  De-reverberation
  Blind Signal Separation (BSS)

  Applications
  VoIP / Telephony
  Tele-conferencing
  Human / machine interface
  Speech based command and control
  Voice annotation
  Audio notes

For more information about acoustic beamforming, check out the following links: