Multimodal Discussion & Engagement for Pitt EDGE

VoiceThread provides four distinct commenting modalities that students and faculty can use interchangeably within any discussion:

Threaded discussions: Comments are organized by slide/media element and can be threaded, allowing focused conversations around specific content. Faculty and students can reply directly to any comment, building on ideas in a structured way.

Asynchronous flexibility: All VoiceThread interactions are asynchronous by design. Students with variable clinical schedules can participate when their time permits. Comments are time-stamped, allowing faculty to track participation patterns.

AI Summaries: VoiceThread automatically generates machine captions for all audio and video content immediately upon creation. For Fall 2026, comprehensive transcript export will enable Pitt to leverage existing AI tools (PittGPT, Microsoft 365 Copilot Chat, Claude for Education, Google NotebookLM) for summarization within platforms already approved for University data.

This approach offers: Compliance alignment with tools Pitt IT has already vetted • No additional approvals needed • Flexibility to apply institution-specific prompts • Cost efficiency by leveraging existing investments

VoiceThread excels at clinical case-based learning—it's a core use case across our health sciences customer base, including programs at Johns Hopkins, Duke, and Mayo Clinic.

Case presentation and reflection: Students upload or record video/audio case reflections directly within VoiceThread. Faculty respond with their own video, audio, or text feedback. The multimodal format captures nuance that text-only tools miss: tone of voice, clinical reasoning process, professional communication skills.

Timestamped and inline annotation: Faculty can leave comments at specific points in student presentations. The animated doodling/annotation tool allows faculty to circle, highlight, or draw attention to specific elements while providing verbal feedback. Unlike any other tool available, commenters have the freedom to roam through content, creating feedback that is simultaneously complex and precise.

Peer learning: Cohort members can view and comment on each other's case reflections with a Student Gallery featureset that has multiple granular policy controls. This peer-to-peer learning mirrors the collaborative case discussion that occurs in clinical team settings.

AI-assisted analysis: Transcripts enable identification of themes, misconceptions, and clinical reasoning patterns using Pitt's institutional AI tools. Faculty can upload VoiceThread transcripts to PittGPT or NotebookLM for secure analysis within Pitt's data environment.

VoiceThread is ideally suited for the hybrid learning model, serving as the connective tissue between asynchronous preparation and synchronous sessions.

Pre-class multimedia prompts: Instructors can create VoiceThreads containing video lectures, slides, documents, images, or any combination of 50+ supported media types. Prompts can include the instructor's own video/audio commentary guiding students through the material. Students engage before class by adding their own comments, questions, and reflections.

Multimodal student responses: Aligning with best accessibility and UDL practices, students choose their preferred modality for responding (video, audio, text, or screen recording). This flexibility accommodates different learning preferences and accessibility needs.

Informing in-class facilitation: Faculty can review student comments before class to identify common questions, misconceptions, or points of interest. Transcripts enable AI-assisted synthesis of student responses using PittGPT or Microsoft Copilot Chat—providing a pre-class summary within tools faculty are already using.

Post-session reinforcement: VoiceThreads remain accessible for student review after synchronous sessions. Faculty can add follow-up comments, additional resources, or responses to questions that emerged during class. The persistent nature of VoiceThread discussions creates a cumulative learning resource.

Sharing multimedia artifacts: Students can upload virtually any project artifact—documents, presentations, images, videos, audio files, PDFs, and more (50+ supported file types). Multiple artifacts of different types can be combined in a single VoiceThread for comprehensive project presentations.

Rich peer feedback: Student Galleries, with granular instructor controls, enable peer feedback in their preferred modality—video, audio, or text. Video and audio feedback is often more nuanced and constructive than text-only comments. The annotation/doodling tool allows peers to point to specific elements while providing verbal feedback. Comment threading enables back-and-forth dialogue between creator and reviewers.

Structured peer review: Faculty can create rubric-aligned prompts guiding peer feedback. VoiceThread assignments integrate with Canvas, allowing structured peer review workflows. Moderation tools allow faculty to review peer feedback before release if desired.

AI-assisted theme identification: Transcripts enable thematic analysis of peer feedback patterns using Pitt's approved AI tools, helping faculty identify common critique patterns and learning opportunities across the cohort.

Accessibility is foundational to VoiceThread's design, not an afterthought.

Modality choice: Students choose how they participate—video, audio, text, or screen recording. This flexibility is core to UDL principles—allowing students to engage through their strengths. Students can switch modalities from comment to comment based on context and preference.

Automatic captioning: All audio and video content is automatically captioned using machine transcription immediately upon creation. No additional steps required from instructors or students—captioning happens fully automatically. Captions display in real-time during playback.

Transcript access and editing: Machine-generated transcripts can be reviewed and edited for accuracy. Transcripts support students who prefer reading to listening, as well as those who are deaf or hard of hearing. Integration with multiple third-party human-powered captioning services is available.

Multilingual support: Automatic captioning supports multiple languages. Interface available in multiple languages.

Accessible navigation: VoiceThread is designed for keyboard navigation and screen reader compatibility. WCAG 2.1 AA compliant. VPAT documentation available upon request.