Our two articles are published in proceedings of INTERSPEECH 2021
INTERSPEECH has grown into the world's largest technical conference focused on speech processing and application with over 1000 attendees and over about 1000 papers. The conferences emphasize interdisciplinary approaches addressing all aspects of speech science and technology, ranging from basic theories to advanced applications.
This year, the 22nd INTERSPEECH 2021 conference was held in a hybrid format in Brno, Czech Republic from August 30 to September 3, 2021.
Our laboratory, together with German colleagues, took part in the annual international competition in computational paralinguistics (Interspeech Computational Paralinguistics ChallengE, ComParE) as the framework of the international conference INTERSPEECH 2021.
This year, the organizers have proposed four sub-challenges:
- the Escalation sub-challenge - a three-way assessment of the level of escalation in a dialogue. In this sub-challenge, we took third place.
- the COVID-19 Cough, COVID-19 Speech sub-challenges - a binary classification on COVID-19 infection based on coughing sounds and speech.
- the Primates sub-challenge - a classification of four species vs background.
As a result of the conference, our two articles were published, one of them jointly with foreign colleagues.
- Verkholyak O., Dresvyanskiy D., Dvoynikova A., Kotov D.O., Ryumina E., Velichko A., Mamontov D., Minker W., Karpov A. Ensemble-within-ensemble classification for escalation prediction from speech // Proceedings of INTERSPEECH-2021, ISCA, 2021. pp. 481-485. DOI: 10.21437/Interspeech.2021-1821
- Ryumina E., Verkholyak O., Karpov A. Annotation Confidence vs. Training Sample Size: Trade-off Solution for Partially-Continuous Categorical Emotion Recognition // Proceedings of INTERSPEECH-2021, ISCA, 2021. pp. 3690-3694. DOI: 10.21437/Interspeech.2021-1636