This repository offers a user friendly interface to deep representation learning, typically applicable to the exploration of animal vocal repertoire.

Scripts used in Lehnhoff et al., 2022.

This repository offers Convolutional Neural Networks (CNN) architectures along with pre-trained weights for the detection of acoustic signals from several cetacean species.

A python interface is also available in order to use them on your own data.

This project contains the public version of the CALOR-DIAL corpus

ARMILU format of the MultiWOZ2.3 corpus

DECODA corpus - distribution of speech transcriptions and annotations of human-human dialogues collected through the ANR DECODA project

ARMILU (Abstract Representation for Multi-Intent spoken Language Understanding) format of the MultiWOZ2.3 corpus

Bundled code for experiments reproducibility

Extended Label Image Annotation Tool called Labelimge "labelImg.py" with variant cross-platforms required at least Python 3.6, opencv, and has been tested with PyQt 4.8, additionally achieves the object Detection & Multi-Tracking.

Pytorch implementation of a CNN detector trained on the Actoustic Trends Blue Fin library dataset

Convolutionnal neural networks for humpback whale vocalisations and delphinid whistles

this is MORSTS with some data for demo

DTOM with some data for test