Mickael Rouvier

Learn More
This paper presents the LIUM open-source speaker diarization toolbox, mostly dedicated to broadcast news. This tool includes both Hierarchical Agglomerative Clustering using well-known measures such as BIC and CLR, and the new ILP clustering algorithm using i-vectors. Diarization systems are tested on the French evaluation data from ESTER, ETAPE and REPERE(More)
We propose to study speaker diarization from a collection of audio documents. The goal is to detect speakers appearing in several shows. In our approach, each show of the collection is processed separately before being processed collectively, to group speakers involved in several shows. Two clustering methods are studied for the overall processing of the(More)
Statistical classifiers operate on features that generally include both useful and useless information. These two types of information are difficult to separate in the feature domain. Recently, a new paradigm based on a Latent Factor Analysis (LFA) proposed a model decomposition into usefull and useless components. This method was successfully applied to(More)
This paper describes the system developed at LIF for the SemEval-2016 evaluation campaign. The goal of Task 4.A was to identify sentiment polarity in tweets. The system extends the Convolutional Neural Networks (CNN) state of the art approach. We initialize the input representations with embeddings trained on different units: lexical, partof-speech, and(More)
Video genre classification is a challenging task in a global context of fast growing video collections available on the Internet. This paper presents a new method for video genre identification by audio analysis. Our approach relies on the combination of low and high level audio features. We investigate the discriminative capacity of features related to(More)
This paper describes the PERCOLATTE participation to MediaEval 2015 task: “Multimodal Person Discovery in Broadcast TV” which requires developing algorithms for unsupervised talking face identification in broadcast news. The proposed approach relies on two identity propagation strategies both based on document chaptering and restricted overlaid names(More)
This paper proposes to learn a set of high-level feature representations through deep learning, referred to as Speaker Embeddings, for speaker diarization. Speaker Embedding features are taken from the hidden layer neuron activations of Deep Neural Networks (DNN), when learned as classifiers to recognize a thousand speaker identities in a training set.(More)
Deep neural networks (DNN) are currently very successful for acoustic modeling in ASR systems. One of the main challenges with DNNs is unsupervised speaker adaptation from an initial speaker clustering, because DNNs have a very large number of parameters. Recently, a method has been proposed to adapt DNNs to speakers by combining speaker-specific(More)