Background noise reduction design for dual microphone cellular phones: Robust approach

Yung Yue Chen, Jia Hao Zhang

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

Cellular phones are used almost everywhere for the communication purpose around the world. However, until now, in noisy environments, the speech communication quality of cellular phones is still awful because the noisy environments where people truly talk are quite complicated and there exists no suitable background noise reduction solution for dealing with this problem effectively. Based on these reasons, this research successfully proposes a novel noise reduction methodology to solve the background noise corrupted problem of cellular phones via a systematic design that comprises a dual microphone array (hardware) and a novel algorithm combined with a background noise whitening process, the system modeling method, and the robust estimation concept (software). After practical experiments, a fact that this proposed method has the immunity to the really noisy environment where cellular phones are used can be found. This background noise reduction design possesses the following impressive advantages: an average of a 20 dB enhanced noise reduction performance (the averaged value of state-of-the-art methods are about 10-15 dB), and an average similarity of 93% of pure speeches can be obtained for the purified speeches by applying this proposed method. Besides, this proposed method can be easily implemented in any of the one shelf or newly developed cellular phones with a really low cost.

Original languageEnglish
Pages (from-to)852-862
Number of pages11
JournalIEEE/ACM Transactions on Audio Speech and Language Processing
Volume25
Issue number4
DOIs
Publication statusPublished - 2017 Apr

All Science Journal Classification (ASJC) codes

  • Computer Science (miscellaneous)
  • Acoustics and Ultrasonics
  • Computational Mathematics
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Background noise reduction design for dual microphone cellular phones: Robust approach'. Together they form a unique fingerprint.

Cite this