Ignacio Viñals
Research Scientist
ViVoLab, Aragón Institute for Engineering Research (I3A)
University of Zaragoza, Spain
Biography
Born in Zaragoza, I finished my studies of Telecommunications Engineering at University of Zaragoza in 2014. Right after finishing my thesis, I was granted with the Vulcanus Scholarship, meaning a year in Japan as intern in Hitachi Central Research Laboratories. Right after my return I joined ViVoLAB group as doctorate scholar. For the following 5 years I have been developing solutions for artificial intelligence, specifically in speaker technologies such as speaker recognition and diarization. I was also invited by Center for Language and Speech Processing (CLSP) at Johns Hopkins University, Baltimore USA, as visiting student in 2017. All the developed solutions have granted many publications as well as good positions in international evaluations such as NIST SRE (2016,2018,2019, 2021), DIHARD (2018, 2019) and Albayzín (2018, 2020) Since 2020 I work in ViVoLAB as speech scientist in the field of speaker recognition and diarization.
Publications
2022
Multimodal Diarization Systems by Training Enrollment Models as Identity Representations Artículo de revista
En: Applied Sciences, vol. 12, no. 3, pp. 1141, 2022.
2021
The Domain Mismatch Problem in the Broadcast Speaker Attribution Task Artículo de revista
En: Applied Sciences, vol. 11, no. 18, pp. 8521, 2021.
ViVoLAB Multimodal Diarization System for RTVE 2020 Challenge Conferencia
Proceedings of XI Jornadas en Tecnología del Habla and VII Iberian SLTech (Iberspeech), vol. 2021-March, 2021.
Diarization and Identity Attribution Compatibility in the Albayzin 2020 Challenge Conferencia
Proceedings of XI Jornadas en Tecnología del Habla and VII Iberian SLTech (Iberspeech), vol. 2021-March, 2021.
2020
Multiclass audio segmentation based on recurrent neural networks for broadcast domain data Artículo de revista
En: EURASIP Journal on Audio, Speech, and Music Processing, vol. 2020, pp. 1–19, 2020.
