The ruffed lemur (Varecia variegata variegata), like all other lemurs, is endemic to Madagascar and inhabits the eastern rainforests of the island. A captive breeding project for this species has been underway since the Sixties and lead to a relatively great population of captive ruffed lemurs. Part of this population was recorded for the purpose of this study. We built a computational model of a non-human primate vocal tract deriving information from a silicon cast of a cadaver and then compared the predicted acoustic response with the formant pattern observed in natural calls. This analysis was performed by considering both oral and nasal tracts. In fact, data about first formants, F1 and F2, derived from the phonetic analysis of V. v. variegata vocalizations, were compared to formant values predicted by the computational model, showing a moderate percentual difference. The vocal repertoire of Varecia sp. features calls showing different formant patterns that can be described by mean simple tube and multiple tube systems. For the first time, this paper provide vocal tract modeling of a prosimian primate and consider variation across different calls. Our findings suggest that if formants in some vocal types can be effectively predicted by mean of a uniform tube model of the oral vocal tract, other vocalizations required the implementation of nasal tract models and multi-tube vocal tract emulations.
Vocal Tract Modeling In A Prosimian Primate: The Black And White Ruffed Lemur.
GAMBA, Marco;GIACOMA, Cristina
2006-01-01
Abstract
The ruffed lemur (Varecia variegata variegata), like all other lemurs, is endemic to Madagascar and inhabits the eastern rainforests of the island. A captive breeding project for this species has been underway since the Sixties and lead to a relatively great population of captive ruffed lemurs. Part of this population was recorded for the purpose of this study. We built a computational model of a non-human primate vocal tract deriving information from a silicon cast of a cadaver and then compared the predicted acoustic response with the formant pattern observed in natural calls. This analysis was performed by considering both oral and nasal tracts. In fact, data about first formants, F1 and F2, derived from the phonetic analysis of V. v. variegata vocalizations, were compared to formant values predicted by the computational model, showing a moderate percentual difference. The vocal repertoire of Varecia sp. features calls showing different formant patterns that can be described by mean simple tube and multiple tube systems. For the first time, this paper provide vocal tract modeling of a prosimian primate and consider variation across different calls. Our findings suggest that if formants in some vocal types can be effectively predicted by mean of a uniform tube model of the oral vocal tract, other vocalizations required the implementation of nasal tract models and multi-tube vocal tract emulations.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.