Education & Resources

Voice conversion (VC) is a significant aspect of artificial intelligence. It is the study of how to convert one’s voice to sound like that of another without changing the linguistic content.

Thanks to their ability to monitor physical activity and health-related parameters, wearable devices are becoming nowadays more and more popular. In addition to what they already offer, an interesting capability achievable through such devices is biometric recognition.

Intelligent reflecting surface (IRS) has recently emerged as a promising paradigm for future wireless communications. Owing to its remarkable capability in reshaping the propagation environment, IRS has a great potential to boost spectral efficiency, mitigate interference, and enhance physical security. 

As the most widely-used spatial filtering approach for multi-channel signal separation, beamforming extracts the target signal arriving from a specific direction. We present an emerging approach based on multi-channel complex spectral mapping, which trains a deep neural network (DNN) to directly estimate the real and imaginary spectrograms of the target signal from those of the multi-channel noisy mixture.