Multimodal Instruction Samples: Expert and Peer
Collection
Sample sets: musical instrument instruction by an expert, and peer collaboration on circuit builds. Public and gated; each has a sign-in viewer. • 4 items • Updated
Access to these samples
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
These recordings are samples shared under the terms set out in the Terms section of this page: evaluation only, no redistribution, no cloning of anyone's voice or likeness. Tell us who you are and how you intend to use them; access is granted at once to the account that agrees.
Log in or Sign Up to review the conditions and access this dataset content.
Multimodal Peer Collaboration Samples - Embodied Map Task with Two Camera Angles
Two non-experts collaborate to build working circuits under asymmetric information: the instructor has the manual, the student has the components, and synchronized audio and dual-camera video capture how shared understanding emerges.
▶ Watch the interactions · See Expert Instruction samples · Discuss the full collection
Sister collection: Expert Instruction, a teacher and a student in structured lessons. Compare peer collaboration under asymmetric information with expert-led instruction.
In this sample: 6 interactions · 41 minutes · 6 circuit projects · 2 studio sessions
Audio: 48 kHz / 32-bit speaker-separated source tracks · sub −60 dB recording noise floor · RT60 < 90 ms
Video: dual 1920 × 1080 / 30 fps camera originals · synchronized front and overhead views
6K source video: available in the full collection · not included in these samples
What makes this collection different?
- Real human collaboration, not generated dialogue. Two non-experts work through each circuit build in an unscripted conversation.
- Asymmetric information creates genuine grounding. The instructor can consult the manual but cannot touch the board; the student can manipulate the components but cannot see the instructions.
- Speech and action share one timeline. Separate voices, a frontal student view, an overhead hands-and-board view, transcripts, and task materials can be analyzed together.
- Interaction context accompanies the recordings. Assigned tasks, participant profiles and project diagrams support analysis of communication, correction, coordination, and outcome.
- Longitudinal change within one pair. The project builds are featured in the same sequence they were attempted within a session. Sequence position is important as participants converge on shared linguistic conventions and become more efficient over time.
View samples
Choose any circuit project to inspect the collaboration on one synchronized timeline:
- Two camera angles: a frontal view of the student and an overhead view of the student's hands, components, and circuit board.
- Separate voices: instructor and student microphone waveforms with independent on/off controls.
- Timed transcript: speaker-labeled, machine-generated turns aligned to playback.
- Task context: the instructions given separately to the instructor and student, plus the project diagram.
- Interaction metadata: project, its place in the session's sequence of builds, duration, session, participant information, and recording specifications.
Transcripts were generated with nvidia/parakeet-tdt-0.6b-v3 and are not human-reviewed.
The six circuit projects
| Project | Duration | Session | Position in session |
|---|---|---|---|
| Electric Light | 3:02 | S021 | build 1 of 6 |
| Flying Saucer | 2:41 | S021 | build 2 of 6 |
| Meet the Music IC | 7:47 | S021 | build 3 of 6 |
| Symphony | 15:51 | S021 | build 4 of 6 |
| Siren | 3:42 | S022 | build 1 of 2 |
| Using Parts as Conductors | 8:03 | S022 | build 2 of 2 |
Builds appear in the order the pair made them; the last column gives each build's place among all the builds of its session, including builds not in this sample. The same two participants exchange instructor and student roles across the two sessions.
Modalities: Instructor microphone · Student microphone · Frontal student camera · Overhead hands-and-board camera · Time-aligned transcripts · Project diagrams
Annotations: Assigned tasks · Participant profiles · Camera offsets and synchronized browser copies
Technical documentation
See TECHNICAL.md for recording formats, camera synchronization, transcript structure, participant profile fields, the complete schema, and loading examples.
Terms
Fluid Concepts provides these recordings solely for evaluation by people who have agreed to these terms on the data page.
- Evaluation only. Use the samples to evaluate the Multimodal Peer Collaboration - Embodied Map Task with Two Camera Angles collection for a possible engagement with Fluid Concepts. Do not use them to train, fine-tune or evaluate a model for any other purpose without written agreement.
- No redistribution. Do not share the files, or derivatives of them, with anyone who has not agreed to these terms on the data page, and do not post them publicly.
- People. The voices and faces belong to paid participants recorded with their consent; the video shows the student. Do not attempt to identify anyone. Participants are identified by studio codes; first names spoken in the sessions are left as recorded.
- No cloning. Do not use the recordings to clone, synthesize or imitate anyone's voice, or to build a digital likeness or avatar of any participant.
- Revocation. Access can be withdrawn at any time. On request, delete all copies.
- No warranty. The samples are provided as they are.
Questions: hello@fluidconcepts.ai
The same text is TERMS.md in this repository (public copy); requesting access is accepting it.
- Downloads last month
- 131