We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

eess.AS

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Electrical Engineering and Systems Science > Audio and Speech Processing

Title: Measuring Mother-Infant Emotions By Audio Sensing

Abstract: It has been suggested in developmental psychology literature that the communication of affect between mothers and their infants correlates with the socioemotional and cognitive development of infants. In this study, we obtained day-long audio recordings of 10 mother-infant pairs in order to study their affect communication in speech with a focus on mother's speech. In order to build a model for speech emotion detection, we used the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) and trained a Convolutional Neural Nets model which is able to classify 6 different emotions at 70% accuracy. We applied our model to mother's speech and found the dominant emotions were angry and sad, which were not true. Based on our own observations, we concluded that emotional speech databases made with the help of actors cannot generalize well to real-life settings, suggesting an active learning or unsupervised approach in the future.
Subjects: Audio and Speech Processing (eess.AS); Machine Learning (cs.LG); Sound (cs.SD); Machine Learning (stat.ML)
Cite as: arXiv:1912.05920 [eess.AS]
  (or arXiv:1912.05920v1 [eess.AS] for this version)

Submission history

From: Xuewen Yao [view email]
[v1] Tue, 10 Dec 2019 19:49:35 GMT (1123kb,D)

Link back to: arXiv, form interface, contact.