RESUMO
A deep learning Phonet model was evaluated as a method to measure lenition. Unlike quantitative acoustic methods, recurrent networks were trained to recognize the posterior probabilities of sonorant and continuant phonological features in a corpus of Argentinian Spanish. When applied to intervocalic and post-nasal voiced and voiceless stops, the approach yielded lenition patterns similar to those previously reported. Further, additional patterns also emerged. The results suggest the validity of the approach as an alternative or addition to quantitative acoustic measures of lenition.
Assuntos
Acústica , Linguística , ProbabilidadeRESUMO
The present study tested whether there is cross-interference between electromagnetic articulography (EMA) and electroglottography (EGG) during the acquisition of kinematic speech data. In experiments 1A and 1B, EMA sensors were calibrated with and without EGG electrodes present in the EMA field. In experiment 2, EMA was used to record lip, tongue, and jaw movements for one male speaker and one female speaker, with and without simultaneous EGG recording. Collectively, the results provide no evidence of signal artifacts in either direction, suggesting that EMA and EGG technology can be combined to reliably assess laryngeal and supralaryngeal motor coordination in speech.