Center for Language & Speech Processing(CLSP), JHU Courses

I-Vector Representation Based on GMM and DNN for Audio Classification

Explore advanced I-vector techniques for audio classification, including GMM and DNN-based approaches. Learn about subspace modeling and applications in language and dialect recognition.

• YouTube
• 57 minutes
• On-Demand
• Free Video

Learning Energy-Based Models of High-Dimensional Data - 2002

Explore energy-based models for high-dimensional data analysis. Learn about deterministic networks, perceptual inference, and efficient learning techniques for large-scale models.

• YouTube
• 1 hour 5 minutes
• On-Demand
• Free Video

Auditory Attention: From Saliency to Models to Applications

Explore auditory attention models and applications, from saliency to speech recognition, understanding human-machine interactions through implicit signals like eye gaze and face pose.

• YouTube
• 1 hour 7 minutes
• On-Demand
• Free Video

Speech Recognition Research: Parallels with Astronomy Development

Explore speech recognition's state-of-the-art, drawing parallels with astronomy's development. Examine cautionary tales from astronomy and biology, encouraging audience participation.

• YouTube
• 45 minutes
• On-Demand
• Free Video

Multilingual Speech to Speech Translation for Under-Resourced Languages

Explore cutting-edge AI techniques for seamless multilingual communication, focusing on under-resourced languages through multi-modal and transfer learning approaches.

• YouTube
• 3 hours 7 minutes
• On-Demand
• Free Video

Bernice: A Multilingual Pre-trained Encoder for Twitter

Explore a multilingual RoBERTa language model trained on 2.5 billion tweets, outperforming adapted models on Twitter benchmarks across languages.

• YouTube
• 24 minutes
• On-Demand
• Free Video

Weak and Strong Learning of Context-Free Grammars - 2012

Explore recent advancements in unsupervised learning of context-free grammars and the transition from weak to strong learning algorithms for accurate syntactic structure prediction.

• YouTube
• 1 hour 3 minutes
• On-Demand
• Free Video

Scalable and Robust Conversational Systems - 2020

Explore recent research on enhancing conversational systems' robustness and scalability through bias analysis, language embeddings, and zero-shot learning techniques.

• YouTube
• 51 minutes
• On-Demand
• Free Video

Multilingual and Code-Switching Speech Recognition

Explore multilingual and code-switching speech recognition techniques, enhancing your understanding of advanced language processing methods.

• YouTube
• 54 minutes
• On-Demand
• Free Video

Multilingual Speech-to-Speech Translation for Under-Resourced Languages

Explore multilingual speech-to-speech translation techniques for under-resourced languages, focusing on encoding, decoding, and analysis strategies.

• YouTube
• 28 minutes
• On-Demand
• Free Video

Leveraging Pre-trained Models for Speech Processing

Explore pre-trained models for speech processing, focusing on leveraging existing knowledge to enhance speech-related tasks and applications.

• YouTube
• 50 minutes
• On-Demand
• Free Video

Diarization in the Wild: Challenges and Opportunities

Explore challenges and opportunities of diarization in daily life recordings, focusing on technical, ethical, and legal aspects for studying spontaneous language use and its applications.

• YouTube
• 59 minutes
• On-Demand
• Free Video

Introduction to Arabic Natural Language Processing - Part 1

Explore Arabic NLP fundamentals, covering linguistic challenges and computational approaches for processing the Arabic language.

• YouTube
• 39 minutes
• On-Demand
• Free Video

Leveraging Pre-training Models for Speech Processing

Explore advanced pre-training techniques for speech processing, including prosody modeling, efficient downstream task adaptation, robustness enhancement, and development of resource-efficient models.

• YouTube
• 2 hours 59 minutes
• On-Demand
• Free Video

Multilingual and Code-Switching Speech Recognition

Explore advanced techniques for multilingual and code-switching speech recognition, focusing on Arabic, English, and low-resource languages. Develop practical LVCSR systems and novel evaluation methods.

• YouTube
• 3 hours 25 minutes
• On-Demand
• Free Video