Speech Recognition Enhancement

Applications  / Smartphones  / Speech Recognition Enhancement








Far-field voice activated devices have become quite popular in the living room. Alango has developed a multi-microphone array technology providing Speech Recognition Enhancement in smartphones integrating 2~4 microphones. Alango’s speech recognition enhancement adapts to different use modes to provide optimal performance while being used as a traditional close talk (handset) phone, handheld or tabletop speakerphone, and in far-field mode up to a few meters away.



Block Diagram

    Dual microphone configuration     3 microphone configuration     4 microphone configuration  





  Acoustic Echo Canceller (AEC)  

Eliminates stereo or mono acoustic echo of voice prompts or music being played by phone in each microphone signal.  Acoustic echo canceller allow the system to respond to the user’s voice during music playback or barge-in.

  Adaptive Microphone Array   A microphone array utilizing a proprietary adaptive beamforming technology that preserves the speech from ambient noise and detects the direction of the voice commands.  





  • Flexible implementation of optimal voice pick up and transmission to the ASR according to the environment and number of microphones.
  • Optimal usage of available resources and minimal power consumption.
  • Portable to all major computational platforms.



Test Results


Comparison of voice trigger recognition rate in a controllable and reproducible acoustic environment (office meeting room), using:

  • 1 microphone raw audio (without any additional processing)
  • 2 microphones processed audio (2mic multidirectional voice enhancement)
  • 3 microphones processed audio (3mic multidirectional voice enhancement)
  • 4 microphones processed audio (4mic multidirectional voice enhancement)




More for Smartphones







Top ▲