Exploring the Potential of Lexical Paraphrases for Mitigating Noise-Induced Comprehension Errors

Listening in noisy environments can be difficult even for individuals with a\nnormal hearing thresholds. The speech signal can be masked by noise, which may\nlead to word misperceptions on the side of the listener, and overall difficulty\nto understand the message. To mitigate hearing difficulties on listeners, a\nco-operative speaker utilizes voice modulation strategies like Lombard speech\nto generate noise-robust utterances, and similar solutions have been developed\nfor speech synthesis systems. In this work, we propose an alternate solution of\nchoosing noise-robust lexical paraphrases to represent an intended meaning. Our\nresults show that lexical paraphrases differ in their intelligibility in noise.\nWe evaluate the intelligibility of synonyms in context and find that choosing a\nlexical unit that is less risky to be misheard than its synonym introduced an\naverage gain in comprehension of 37% at SNR -5 dB and 21% at SNR 0 dB for\nbabble noise.\n