no code implementations • 8 Jan 2024 • Jennifer Williams, Karla Pizzi, Paul-Gauthier Noe, Sneha Das
Most recent speech privacy efforts have focused on anonymizing acoustic speaker attributes but there has not been as much research into protecting information from speech content.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +2
no code implementations • 2 Jun 2023 • Angélica S. Z. Suárez, Clément Laroche, Line H. Clemmensen, Sneha Das
The evaluation of such algorithms often relies on reference-based objective metrics that are shown to correlate poorly with human perception.
no code implementations • 27 Apr 2023 • Laurits Fromberg, Sneha Das, Line Katrine Harder Clemmensen
Our method achieves better results for most time features as well as for a subset of the frequency features.
no code implementations • 11 Jul 2022 • Flavia D. Frumosu, Nicole N. Lønfeldt, A. -R. Cecilie Mora-Jensen, Sneha Das, Nicklas Leander Lund, A. Katrine Pagsberg, Line K. H. Clemmensen
However, coding human behavior is a time-consuming, expensive task, in which reliability can be difficult to achieve and bias is a risk.
no code implementations • 11 May 2022 • Nicole N. Lønfeldt, Flavia D. Frumosu, A. -R. Cecilie Mora-Jensen, Nicklas Leander Lund, Sneha Das, A. Katrine Pagsberg, Line K. H. Clemmensen
Features from the videos will be extracted and used to compute ratings of behavior, which will be compared to ratings of behavior produced by mental health professionals trained to use a specific behavioral coding manual.
no code implementations • 25 Apr 2022 • Sneha Das, Nicole Nadine Lønfeldt, Anne Katrine Pagsberg, Line. H. Clemmensen
Through our work in this paper, we learned that the model with default classification threshold performs worse on children from the patient group.
no code implementations • 28 Mar 2022 • Sneha Das, Nicklas Leander Lund, Nicole Nadine Lønfeldt, Anne Katrine Pagsberg, Line H. Clemmensen
Furthermore, to address the lack of activation and valence labels in the transfer datasets, we annotate the signal samples with activation and valence levels corresponding to a dimensional model of emotions, which were then used to evaluate the quality of the embedding over the transfer datasets.
no code implementations • 28 Mar 2022 • Sneha Das, Nicole Nadine Lønfeldt, Anne Katrine Pagsberg, Line H. Clemmensen
We show that while the DAE has the highest classification accuracy among the methods, the semi-supervised VAE has a comparable classification accuracy and a more consistent latent embedding distribution over data sets.
no code implementations • 5 May 2021 • Sneha Das, Nicole Nadine Lønfeldt, Anne Katrine Pagsberg, Line H. Clemmensen
Furthermore, due to the black-box nature of deep learning algorithms, a newer challenge is the lack of interpretation and transparency in the models and the decision making process.