publication . Article . 2020

Voice activity detection in noisy conditions using tiny convolutional neural network

Open Access Russian
  • Published: 01 Jun 2020 Journal: Informatika, volume 17, issue 2, pages 36-43 (issn: 1816-0301, Copyright policy)
  • Publisher: The United Institute of Informatics Problems of the National Academy of Sciences of Belarus
Abstract
<jats:p>The paper investigates the problem of voice activity detection from a noisy sound signal. An extremely compact convolutional neural network is proposed. The model has only 385 trainable parameters. Proposed model doesn’t require a lot of computational resources that allows to use it as part of the “internet of things” concept for compact low power devices. At the same time the model provides state of the art results in voice activity detection in terms of detection accuracy. The properties of the model are achieved by using a special convolutional layer that considers the harmonic structure of vocal speech. This layer also eliminates redundancy of the mo...
Persistent Identifiers
Subjects
free text keywords: voice activity detector, harmonic signal, convolutional neural network, pitch, speech processing, lcsh:Electronic computers. Computer science, lcsh:QA75.5-76.95, Redundancy (engineering), Voice activity detection, WebRTC, Fundamental frequency, Invariant (physics), Audio signal, Speech recognition, Computer science
Any information missing or wrong?Report an Issue