Speech Technologies Proved Themselves, Even in Extremely Noisy Conditions

The prestigious Journal of the Acoustical Society of America has just published a scientific article evaluating the performance of Phonexia speech technologies in the most difficult conditions. The evaluations were undertaken in extremely noisy environments where the speech itself was barely audible.

Phonexia technologies were able to identify the language and gender of the speaker, identify the speaker themselves, and separate the speakers from one another. A team of researchers was formed that included those from Phonexia, Brno University of Technology, and US commercial and governmental partners.

The team was able to prove that speech technologies can also be successfully applied in such extremely difficult conditions. You can download the full article ”Noise-robust speech triage” here.

Phonexia’s CTO Petr Schwarz was one of the co-authors of the scientific article published in the Journal of the Acoustical Society of America.

Copyright 2018 Acoustical Society of America. This article may be downloaded for personal use only. Any other use requires prior permission of the author and the Acoustical Society of America.
This article appeared in The Journal of the Acoustical Society of America (Vol.143, No.4) and may be found at https://asa.scitation.org/doi/abs/10.1121/1.5031029