1 code implementation • 27 Oct 2022 • Kexin Feng, Theodora Chaspari
Following that, the depression is classified at the global-level from a group of vowel CNN embeddings that serve as the input of another 1D CNN ("depression CNN").
1 code implementation • 27 Oct 2022 • Kexin Feng, Jacqueline B. Duong, Kayla E. Carta, Sierra Walters, Gayla Margolin, Adela C. Timmons, Theodora Chaspari
The proposed metric learning is based on a Siamese neural network (SNN) that learns the relative difference between pairs of samples from a target user and non-target users, thus being able to address the scarcity of labelled data from the target.
1 code implementation • 5 Oct 2022 • Kexin Feng, Theodora Chaspari
Explainability of the high-level information capturing the segment-by-segment decisions is further inspected for participants with and without depression.
no code implementations • 23 Jun 2022 • Zepeng Huo, Bobak J. Mortazavi, Theodora Chaspari, Nicolaas Deutz, Laura Ruebush, Ricardo Gutierrez-Osuna
We built a multitask neural network to estimate the macronutrient composition from the CGM signal, and compared it against a baseline linear regression.
no code implementations • 7 Sep 2021 • Kexin Feng, Theodora Chaspari
Leveraging the abundance of labelled speech data from acted emotions, this paper proposes a few-shot learning approach for automatically recognizing emotion in spontaneous speech from a small number of labelled samples.
no code implementations • 4 Jun 2020 • Kexin Feng, Theodora Chaspari
Our system use samples from the source data to pre-train the weights of proposed Siamese neural network, which are fine-tuned based on the target data.
no code implementations • 16 Jan 2020 • Vansh Narula, Zhangyang, Wang, Theodora Chaspari
Image and video-capturing technologies have permeated our every-day life.