Despite advances in neural language modeling, obtaining a good model on a large scale
multi-domain dataset still remains a difficult task. We propose training methods for building …

Deep learning has brought considerable changes and created a new paradigm in many
research areas, including computer vision, speech processing, and natural language …

This research addresses the language model (LM) domain mismatch problem in automatic
speech recognition (ASR) systems. The ASR systems rely on LMs to constrain its recognition …

Cette thèse s' inscrit dans le cadre d'une étude sur le potentiel de la transcription
automatique pour l'instrumentation de situations pédagogiques. Notre contribution porte sur …

Abstract Language models are a key component of automatic speech recognition systems.
In recent years, language models based on neural networks and in particular recurrent …