Speech communication devices may reproduce non-speech sounds with a loudness exceeding the loudness of the speech they reproduce; this may result in listening discomfort or injury to the listener. To address this issue a novel method of sound amplitude limiting for signals conveying speech has been developed. The method continuously estimates the loudness of speech conveyed by the signal on a frequency specific basis. From these estimates it produces a set of time-varying speech reference levels. It limits the level of sounds conveyed by the signal to these speech reference levels. The method is called speech referenced limiting (SRL). An experiment was conducted in which 16 normal-hearing subjects compared unprocessed and SRL processed telephone speech and higher level non-speech signals. Results show the loudness of SRL processed non-speech signals to be reduced to the loudness of the speech and the quality and loudness of the speech to be preserved.
|Number of pages||4|
|Journal||Proceedings of Forum Acusticum|
|Publication status||Published - 2011|