14thAnnual Conference of the International Speech Communication Association

Lyon, France
August 25-29, 2013

Improving Speech Intelligibility in Noise by SII-Dependent Preprocessing Using Frequency-Dependent Amplification and Dynamic Range Compression

Henning Schepker (1), Jan Rennies (2), Simon Doclo (1)

(1) Carl von Ossietzky Universität Oldenburg, Germany
(2) Fraunhofer IDMT, Germany

In this contribution, a new preprocessing algorithm to improve speech intelligibility in noise is proposed, which maintains the signal power before and after processing. The proposed AdaptDRC algorithm consists of two time- and frequency-dependent stages, which are both functions of the estimated SII. The first stage applies a time- and frequency-dependent amplification, while the second stage applies a time- and frequency-dependent dynamic range compression (DRC). Experiments with a competing speaker (CS) and a speech-shaped noise (SSN) show an increase in speech intelligibility for a wide range of SNRs for four different objective measures that are correlated with speech intelligibility. Listening tests conducted within the framework of the Hurricane Challenge with 175 subjects confirm these findings and show improvements of up to 20.5% in intelligibility for SSN and 12.3% for CS.

Full Paper

Bibliographic reference.  Schepker, Henning / Rennies, Jan / Doclo, Simon (2013): "Improving speech intelligibility in noise by SII-dependent preprocessing using frequency-dependent amplification and dynamic range compression", In INTERSPEECH-2013, 3577-3581.