CCFP — Original TCF preparation audio, model 01 Generated 25 October 2026. These are synthetic training voices, not official TCF recordings. The approved scripts and questions are original CCFP content. Only pauses and audio normalization were added. Synthesis engine: Piper 1.3.0, Open Home Foundation / Michael Hansen and contributors. https://github.com/OHF-Voice/piper1-gpl (GPL-3.0-or-later). The engine is used offline and is not included in this website. Voice models: rhasspy/piper-voices, fr_FR-siwis-medium and fr_FR-mls-medium. https://huggingface.co/rhasspy/piper-voices/tree/main/fr/fr_FR/siwis/medium https://huggingface.co/rhasspy/piper-voices/tree/main/fr/fr_FR/mls/medium The respective model cards identify CC BY 4.0 datasets. License: https://creativecommons.org/licenses/by/4.0/ SIWIS: Yamagishi, Junichi; Honnet, Pierre-Edouard; Garner, Philip; Lazaridis, Alexandros (2017). The SIWIS French Speech Synthesis Database, 2016 [dataset]. University of Edinburgh. https://doi.org/10.7488/ds/1705 Piper SIWIS medium model fine-tuned from the English lessac medium model. MLS: Vineel Pratap, Qiantong Xu, Anuroop Sriram, Gabriel Synnaeve and Ronan Collobert (2020). MLS: A Large-Scale Multilingual Dataset for Speech Research. Read speech from LibriVox audiobooks. https://www.openslr.org/94/ — https://arxiv.org/abs/2012.03411 Piper MLS medium model, speaker indexes 0 and 2. Changes: the models synthesize new CCFP scripts; output normalized and encoded as mono MP3. No real person's endorsement, participation or official exam affiliation is implied.