CoBi's speech-based pipeline is technically feasible in noisy middle school classrooms, with visualizations updating in 10-second intervals¶

Claim · All claims
Evidence · 1 study (2 entries) · design r3 · q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person · recheck by 2028: tool and model accuracy changes quickly

Subclaims¶

q2 i? In three classrooms, browser interfaces loaded properly and visualizations updated as expected in 10-s intervals, reaching end states after about 15 minutes. → Breideband 2025 q2 i? Mismatched time-slider states between teacher and student interfaces caused confusion, leading some students to mistake the class-wide visualization for a group-level one. → Breideband 2025 (2)

Evidence¶

Breideband 2025¶

Breideband, T., Bush, J. B., Reitman, J. G., Rose, S., Weatherley, J., Penuel, W. R., & D'Mello, S. (2025). A Feasibility and Implementation Integrity Study of the Community Builder (CoBi): An AI-based Collaboration Support System in K-12 Classrooms. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-025-00509-2

q2 · i? · design · r3

Classroom pilot of CoBi v1 in three middle school classrooms with 24 fully consented students: "From a technical perspective, CoBi performed adequately in the classrooms." Visualizations reached end states approximately 15 minutes into collaborative activity.

"From a technical perspective, CoBi performed adequately in the classrooms. The browser interfaces (for both CoBi and the audio-visual recorder) loaded properly, and the visualizations updated as expected in 10-s intervals."

Breideband 2025 (2)¶

Breideband, T., Bush, J. B., Reitman, J. G., Rose, S., Weatherley, J., Penuel, W. R., & D'Mello, S. (2025). A Feasibility and Implementation Integrity Study of the Community Builder (CoBi): An AI-based Collaboration Support System in K-12 Classrooms. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-025-00509-2

q2 · i? · design · r3

Field notes and classroom video from the same pilot showed that "there were situations where the state of the visualization on the teacher TV did not match the state of the visualization on students’ Chromebooks," causing students to misread the class-wide display as group-level.

"there were situations where the state of the visualization on the teacher TV did not match the state of the visualization on students’ Chromebooks."

Discussion¶

Related Claims¶

• Middle school students and teachers broadly buy in to an AI system that visualizes classroom community agreements — related
• High-integrity CoBi implementation requires significant teacher professional learning, and teachers adapt the system in diverse ways — related
• Students raise privacy concerns about classroom speech data while teachers initially interpret CoBi as a performance-monitoring device — related
• Students strongly prefer qualitative tree visualizations over quantitative radar charts of classroom collaboration — related
• Speech detection showed preliminary success but remained challenging in noisy settings, particularly for speaker differentiation, children's voices, and diverse dialects — related