Reproducibility in speech rate convergence experiments

Abstract The reproducibility of scientific studies grounded on language corpora requires approaching each step carefully, from data selection and pre-processing to significance testing. In this paper, we report our reproduction a recent study based well-known conversational corpus (Switchboard). reproduced Cohen Priva et al. (J Acoust Soc Am 141(5):2989–2996, 2017) focuses speech rate convergence between speakers in conversation. While confirms the main result original study, it also shows interesting variations details. addition, tested for robustness its pre-processing, as well underlying model rate, variable observed. Our analysis that another approach is needed take into account complex aspects conversations. Another benefit reproducing previous further, testing strengthening results other research teams increasing validity visibility results. line, created notebook scripts which available online.