ViVoLab research group participates at iberSPEECH 2022 conference in Granada

18 Novenber 2022

This week a group of ViVoLab researchers is in Granada, participating in the iberSPEECH 2022 conference. This event held every two years is a meeting point to present and discuss the results of the research on speech and language technologies on Iberian languages and promoting industry/university collaboration.

ViVoLab contributes to the scientific program of the conference with 3 regular papers, covering areas ranging from audio segmentation to speech disorder detection.

• A Study on the Use of wav2vec Representations for Multiclass Audio Segmentation
• Cross-Corpus Speech Emotion Recognition with HuBERT Self-Supervised Representation
• S3prl-Disorder: Open-Source Voice Disorder Detection System based in the Framework of S3PRL-toolkit

Vivolab team also participated in the Albayzín 2022 evaluation challenge, submitting a system for the speech to text track that obtained competitive results:

• ViVoLAB System Description for the S2TC IberSPEECH-RTVE 2022 challenge

Furthermore, ViVoLab was awarded with 2 prizes at the conference. First, the work developed by researcher Victoria Mingote during the last 4 years was recognized by the organizing committee, granting her with the best PhD thesis award:

• Representation and Metric Learning Advances for Deep Neural Network Face and Speaker Biometric Systems

For further information check the conference webpage: Iberspeech 2022