Platform Configuration
System Settings & Evaluative Rubric Calibration
Adjust reproducible rubric scoring weights, configure runtime demo modes, and inspect external API integration health.
1. Reproducible Rubric Weights
Sum: 100% (Target: 100%)
Delivery Stability (Pace bounded, pause quality)20%
Clarity & Precision (Filler density, articulation)20%
Rhetorical Structure (Thesis, transitions, wrap-up)20%
Content Substance (Empirical grounding, density)15%
Fluency & Cadence (Zero hesitations or stalls)15%
Vocal Engagement (Prosodic dynamism, energy)10%
2. Demo Mode & Autonomous Fallbacks
Autonomous Demo Mode (Mandatory Zero-Key Operation)
When active, the platform functions autonomously using seeded speech recordings, deterministic metrics, and offline semantic evaluation engines. No external API keys are required for judging.
3. External Inference API Status
Google Gemini APISemantic & Structural Evaluation
OFFLINE FALLBACKWhisper / Transcription APIAcoustic Phonetic Transcription
LOCAL CORPUS