CLSP Summer Program: Closing Presentations

CLSP Summer Program: Closing Presentations

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

Humanities, Social Sciences & Thought AI — News & Applications UYComputer scienceUYQArtificial intelligence (AI)
🎙 Center for Language & Speech Processing (CLSP), JHU 👥 4K 📅 August 6, 2026 ⏱ 509 min 👁 391 📄 documentary 🧭 2026-08-16
Available in: English (current) Français

Keywords

workshopspeech processingAIhealth coachingmultimodal

Summary

This video is a recording of the closing presentations of the 12th Frederick Jelinek Memorial Workshop, held at Johns Hopkins University. The workshop brought together researchers from academia and industry to work on three projects over six weeks. The video begins with an introduction by Sanjiv Kudanur, who provides an overview of the workshop’s history, organization, and logistics. He explains the process of selecting projects, which involved interactive peer review and voting. The three projects are: building omnimodal encoders, creating a simulator for evaluating spoken conversations, and developing AI-based health coaches. The video includes details about the participants, such as their institutional diversity and academic backgrounds. It also highlights the summer school and plenary lectures that were part of the program. The main portion of the video consists of presentations by the project teams, summarizing their work and findings. The video is long (over 8 hours) and is primarily aimed at an academic audience interested in speech and language processing.

160 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides valuable insights into the organization and outcomes of a collaborative research workshop. The argumentation is based on the experiences and results of the participants, which adds credibility. However, the technical depth is limited, as the presentations are summaries rather than detailed expositions. The value lies in showcasing the collaborative process and the range of topics addressed.

Scientific Rigor, Source Quality, Title Accuracy

The video is a primary source, as it is a recording of the actual presentations. The information is presented by the researchers themselves, which enhances its reliability. However, no external sources are cited, and the content is not peer-reviewed. The title accurately reflects the content, as it is indeed a recording of closing presentations.

128 words

Title / Content Match

The title accurately reflects the content, which consists of closing presentations from the CLSP Summer Program.

Quality & Reliability

7/10

The video is a recording of a workshop's closing presentations, featuring talks by researchers and students. The content is primarily descriptive and organizational, with some technical depth in the project summaries. The information is reliable as it comes from the organizers and presenters, but it is not peer-reviewed and lacks detailed technical exposition.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Contribution & Novelties

The video provides an overview of the latest research directions in speech and language processing, particularly in the areas of multimodal AI, conversational systems, and AI health coaching. It highlights the collaborative nature of research and the importance of interdisciplinary teams. The ‘Pour aller plus loin’ section suggests further exploration of these topics.

Pour aller plus loin :

  • Multimodal learning — Overview of multimodal learning, relevant to the omnimodal encoders project.
  • Spoken dialogue systems — Background on spoken dialogue systems, relevant to the conversation simulator project.
  • AI in healthcare — Overview of AI applications in healthcare, relevant to the health coaching project.

102 words

Radar Profile

The radar profile shows moderate scores across all dimensions, indicating a balanced but not exceptional video. The highest score is in fiabilite_globale, reflecting the credibility of the source, while the lowest is in niveau_technique, as the content is not highly technical.

Reliability 7/10