About me

Postdoctoral Research Fellow, Tampere University

I am currently working with Professor Tuomas Virtanen, group leader of the Audio Research Group at Tampere university. We are currently working on Zero-shot Active learning of acoustic models, a Research Council of Finland funded project.

My research interests include audio signal processing, machine learning, acoustic scene classification, sound event detection and audio captioning.

My career

For a brief and compact overview, take a look at my 2-page CV. I have been working as a postdoctoral research fellow at the Machine Listening Group with assistant professor Annamaria Mesaros, at Tampere University between 2020 and 2024. Between February and August of 2020 I was a Postdoc in the Image and Signal Processing group. I received my PhD in 2019 from the Universitat de València, where I worked in the signal processing and acoustic technology group. I have worked on acoustic signal processing, more specifically in the field of sound event recognition, the topic of my PhD thesis.

I have diverse research interests and I am always happy to dive into a new area.

Contact

Tampere University
Korkeakoulunkatu 7
33720 Tampere
Finland

Office: Korkeakoulunkatu 1, Tietotalo building F419
Email: irene.martinmorato[at]tuni.fi

Datasets

  • AVCaps: An audio-visual dataset with modality-specific captions, contains modality specific captions from 2061 video clips spanning a total of 28.8 hours.
  • MAESTRO Real: Multiple Annotator Estimated STROng labels, contains 49 audio files from 5 different real acoustic scenes, each of them from 3 to 5 min long.
  • MAESTRO Synthetic: Multiple Annotator Estimated STROng labels, contains 20 audio files created using Scaper, each of them 3 min long.
  • MACS: Multi-Annotator Captioned Soundscapes, contains audio captions and corresponding audio tags for a number of 3930 audio files of the TAU Urban Acoustic Scenes 2019 development dataset (airport, public square and park)
  • MATS: Multi-Annotator Tagged Soundscapes, contains audio tags from 133 annotators for a number of 3930 audio files of the TAU Urban Acoustic Scenes 2019 development dataset (airport, public square and park)