Settings & Integrations

Configure AI pipeline and defaults

Model Orchestration Defaults

Configure which AI provider handles specific stages of the meeting minutes pipeline.

Primary Report Generation (Compliance & Structure)

Used for generating the final PROCÈS-VERBAL OFFICIEL.

Speaker Analysis & Extraction

Used to identify talk times and strategic positioning.

Executive Summary & Polish

Used for generating fast, legally-worded summaries.

Transcript Processing

Advanced settings for handling massive audio transcripts.

Context Window Chunk Size (Tokens)

Recommended: 4000-6000 tokens for optimal processing.