Answers are grounded in this paper and cite their sources.
The effect of several factors upon voice identification was examined. These factors were: the size of the class of possible voices, the duration of the speech signal, the frequency range of the speech signal, voicing vs nonvoicing speech characteristics, and the simultaneous presentation of several voices. One of the most effective factors for speaker identification was the duration of the speech signal. Duration, as such, appears to be important, however, only insofar as it admits a smaller or larger statistical sampling of the speaker's speech repertoire.
No takes yet. Share an insight, caveat, or question.
Pollack et al. (1954) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: