Sahar Ghannay

Latest

• Continual self-supervised domain adaptation for end-to-end speaker diarization
• Benchmarking Transformers-based models on French Spoken Language Understanding tasks
• Étude comparative de modèles Transformers en compréhension de la parole en français
• OVERLAP-AWARE LOW-LATENCY ONLINE SPEAKER DIARIZATION BASED ON END-TO-END LOCAL SEGMENTATION
• Evaluating the carbon footprint of NLP methods: a survey and analysis of existing tools
• Where Are We in Semantic Concept Extraction for Spoken Language Understanding?
• Neural Networks approaches focused on French Spoken Language Understanding: application to the MEDIA Evaluation Task
• Error Analysis Applied to End-to-End Spoken Language Understanding
• What is best for spoken language understanding: small but task-dependant embeddings or huge but out-of-domain embeddings?
• A Comparison of Metric Learning Loss Functions for End-To-End Speaker Verification
• A Metric Learning Approach to Misogyny Categorization
• A study of continuous space word and sentence representations applied to ASR error detection
• End-to-end named entity and semantic concept extraction from speech
• TED-LIUM 3: Twice as Much Data and Corpus Repartition for Experiments on Speaker Adaptation
• Task Specific Sentence Embeddings for ASR Error Detection
• Représentations de phrases dans un espace continu spécifiques à la tâche de détection d'erreurs
• Simulation d'erreurs de reconnaissance automatique dans un cadre de compréhension de la parole
• End-to-end named entity extraction from speech
• Simulating ASR errors for training SLU systems
• Enriching confusion networks for post-processing
• ASR Error Management for Improving Spoken Language Understanding
• Using Hypothesis Selection Based Features for Confusion Network MT System Combination
• Acoustic word embeddings for ASR error detection
• Recent improvements on error detection for automatic speech recognition
• Evaluation of acoustic word embeddings
• Utilisation des représentations continues des mots et des paramètres prosodiques pour la détection d'erreurs dans les transcriptions automatiques de la parole
• Word embedding evaluation and combination
• Combining continous word representation and prosodic features for ASR error prediction
• Which ASR errors are hard to detect?
• Word embeddings combination and neural networks for robustness in ASR error detection