Extracting music and noise from the ear canal using a headset

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Researchers

Research units

  • Nokia

Abstract

An adaptive real-time filtering technique for separating music and ambient noise signals in the ear canal of a headset user is proposed. The system has been tested with a dummy head under laboratory conditions. This paper compares the responses given by the real-time system with those obtained with standard laboratory measurements. Both the headphone response estimate and the noise level measurements remain highly accurate even at a high ambient noise level of 86 dBSPL inside the blocked ear canal. An efficient multirate version of the method is proposed for cases, where the ambient noise is dominated by low frequencies. The method has several useful applications for music listening in noise, including adaptive headphone target response equalization and unmasking of music.

Details

Original languageEnglish
Title of host publicationProceedings of the AES International Conference
Subtitle of host publication2019 AES International Conference on Headphone Technology
Publication statusPublished - 21 Aug 2019
MoE publication typeA4 Article in a conference publication
EventAES International Conference on Headphone Technology - San Francisco, United States
Duration: 27 Aug 201929 Aug 2019

Conference

ConferenceAES International Conference on Headphone Technology
CountryUnited States
CitySan Francisco
Period27/08/201929/08/2019

    Research areas

  • Acoustics, adaptive filtering, audio systems, headphones

ID: 38782133