French Full-Duplex Conversation Dataset

Overview

Naturalistic, two-speaker French conversations captured at studio quality in full-duplex stereo. Pairs of native French speakers from metropolitan France, Belgium, and French-speaking parts of Canada discuss everyday topics for the full duration of the session — no read scripts, no scene cuts. Each recording preserves real overlapping speech, backchannels, hesitations, and code-switching, so downstream models train on the way French actually sounds in the wild. Every clip is collected from paid contributors with explicit consent, scene-level provenance, and metadata for speaker demographics, dialect, and acoustic environment.

Key highlights

• 01

  Hexagonal France, Belgian, and Québécois pairings with per-speaker dialect tags so models can learn regional pronunciation drift.

• 02

  Verlan, register switches between formal vous and informal tu, and Parisian abbreviations ("chui", "ouais") preserved as-spoken.

• 03

  Café-style conversational cadence — interruptions, agreement particles ("voilà", "bah"), and rhetorical questions captured intact.

• 04

  Regional vocabulary and pronunciation differences across Lyon, Marseille, Brussels, and Montréal annotated in the speaker metadata.

Technical specifications