The effect of interacting with different interlocutors on intraspeaker speech variability is researched since the 1970s. While the traditional position of the Speech Accommodation Theory argued that speakers accommodate their speech features according to their changing social goals in interaction, more recent predictive code models such as the Interactive Alignment Model assume that speakers converge to their interlocutors over time to facilitate cognitive processing of speech in interaction. The two positions make different assumptions regarding the nature of and the motivation underlying interlocutor-dependent intraspeaker variability. Nevertheless, researchers of phonetic convergence in interaction use mostly the same designs to collect evidence for both theories. The current paper presents a study designed specifically to test one of the hypotheses of the Interactive Alignment Model. A single female speaker of Modern Israeli Hebrew was recorded in dyadic interactions with four close acquaintances. 20 utterances were extracted for each speaker from each recording, controlling for syllable count. A positive correlation was found between the speech rate of the observed single speaker and that of her interlocutors in each individual interaction, supporting the Interactive Alignment Model. Simultaneously, the results point to possible integration of social parameters affecting speech variability on the local level of conversation.
Cite as: Kalmanovitch, Y. (2020) Interlocutor-dependent intra-speaker speech rate variability in interaction: a pilot study on four conversations in modern Hebrew. Proc. Speech Prosody 2020, 319-323, doi: 10.21437/SpeechProsody.2020-65
@inproceedings{kalmanovitch20_speechprosody,
author={Yshai Kalmanovitch},
title={{Interlocutor-dependent intra-speaker speech rate variability in interaction: a pilot study on four conversations in modern Hebrew}},
year=2020,
booktitle={Proc. Speech Prosody 2020},
pages={319--323},
doi={10.21437/SpeechProsody.2020-65}
}