Solution: Microphone Array & Voice Interfaces

Capturing clean human voices in noisy environments (such as those in kitchens, living rooms, or gyms) is a critical issue that must be addressed for smart devices like TVs, set-top boxes, and Soundbars to perform voice interactions and voice calls effectively. In real life situations, various types of noise in noisy environments hinder the effective capture of human voices by such devices. These noises include:

  • Sounds played by the device itself, such as music being played by a TV, set-top box, or Soundbar
  • Ambient steady and non-steady, diffuse noises like those from fans or air conditioners
  • Point noises within the environment, such as noises emitted by a TV placed at a specific location

In addition to these noises, the excessively loud playback from SoundBars or TVs also makes it difficult to accurately and effectively capture useful speaker voices.

A high-performance voice interface solution is particularly important in these devices. A high-performance voice interface solution, in addition to tackling disturbing noise issues, also provides long-distance pickup and voice interruption (Barge-in). Such a front-end voice scheme can then output clean and effective human voices for voice interaction (ASR) and conference calls (Communication).