“Young” and “Old” Voice: the prosodic auto-transplantation technique for speaker’s age recognition

Massimo Pettorino, Elisa Pellegrino, Marta Maffia


The present study is intended to figure out the extent to which prosody and intonation entail listeners’ ability to estimate the speaker’s age. The performance of a 40-year old anchorman and that produced by the same speaker at the age of 80 were spectro-acoustically analyzed in order to identify the prosodic features of the “young” and the “old” voice. The results of the analyses have shown relevant differences between the two voices on suprasegmental level. To test the effects of these differences on perceptual level, through the prosodic transplantation technique, the F0 values and the durations of segments and silences were transferred from the “young” to the “old” voice and viceversa. Two age recognition tests, based on original and transplanted voices, were administered to Italian listeners. The results of perceptual tests have confirmed the strict relationship between some rhythmic and prosodic features and the speaker’s age and have demonstrated the effectiveness of the transplantation technique. With advancing age, articulation rate and speech rate slow down, voice register raises and tonal range widens. Moreover, the “old” voice is also characterized by a higher percentage of vocalic portion that determines a shift of Italian rhythm towards the isomoraic pattern


 DOI: 10.21437/SpeechProsody.2014-15

Cite as: Pettorino, M., Pellegrino, E., Maffia, M. (2014) “Young” and “Old” Voice: the prosodic auto-transplantation technique for speaker’s age recognition. Proc. 7th International Conference on Speech Prosody 2014, 135-139, DOI: 10.21437/SpeechProsody.2014-15.


@inproceedings{Pettorino2014,
  author={Massimo Pettorino and Elisa Pellegrino and Marta Maffia},
  title={{“Young” and “Old” Voice: the prosodic auto-transplantation technique for speaker’s age recognition}},
  year=2014,
  booktitle={Proc. 7th International Conference on Speech Prosody 2014},
  pages={135--139},
  doi={10.21437/SpeechProsody.2014-15},
  url={http://dx.doi.org/10.21437/SpeechProsody.2014-15}
}