Early identification of cognitive or physical overload is critical in fields where human decision-making matters when preventing safety and property threats. Pilots, drivers, surgeons, and nuclear plant operators are among those affected by this challenge since acute stress may impair their cognition. In such a context, the significance of paralinguistic automatic speech processing rises for early stress detection. The intensity, intonation, and cadence of an utterance are examples of paralinguistic traits that determine the meaning of a sentence and are often lost in the verbatim transcript. To address this issue, tools to recognize paralinguistic traits effectively are constantly being developed. However, a data bottleneck still exists in the training of paralinguistic speech traits, and the lack of high-quality reference data for the training of artificial systems persists. Regarding this, we present an original empirical dataset collected using the experimental protocol BESST for capturing speech signals provided under induced stress. With this data, we aim to promote the development of pre-emptive intervention systems based on stress estimation from speech.
During the development of the BESST dataset, we have created two software tools.
Re-synchronization tool for recovering ECG synchronization after suffering from linear time drift. The code for the re-synchronization can be found at https://github.com/BUTSpeechFIT/besst-tools.
To create correct phase alignments, we created BESST ANNOtation tool - BESSTiANNO.