{"id":"openbmb-voxcpm2","nom":"VoxCPM2","editeur":"OpenBMB","pays":"Chine","poids_ouverts":true,"poids_selon":null,"licence":"Apache 2.0","licence_selon":null,"liens":{"page":"https://huggingface.co/openbmb/VoxCPM2","essayer":"https://huggingface.co/spaces/OpenBMB/VoxCPM-Demo"},"fiche_auto":true,"page":"https://quelleia.com/voix/modeles/openbmb-voxcpm2/","releve":"2026-10-19","en_bref":"5e sur 16 en prononciation du français.","presentation":"Modèle de synthèse vocale d'OpenBMB sans tokeniseur audio, à architecture autorégressive par diffusion, qui produit un son en 48 kHz dans 30 langues. Il crée une voix à partir d'une simple description textuelle et clone une voix à partir d'un court extrait, avec un contrôle optionnel du style.","sortie":"2026-04","taille":"2 milliards de paramètres","langues":{"annoncees":{"annoncees":"30 langues, dont le français","nombre":30,"francais":true,"liste":["ar","my","zh","da","nl","en","fi","fr","de","el","he","hi","id","it","ja","km","ko","lo","ms","no","pl","pt","ru","es","sw","sv","tl","th","tr","vi"]},"prises_en_charge_selon_open_tts":30,"evaluees":["fr"]},"preuves":[{"appuie":["langues","presentation","taille"],"lu":"https://huggingface.co/openbmb/VoxCPM2 (« VoxCPM2 is a tokenizer-free, diffusion autoregressive Text-to-Speech model - 2B parameters , 30 languages , 48kHz audio output »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/openbmb/VoxCPM2 (« Voice Design - Generate a novel voice from a natural-language description alone »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/openbmb/VoxCPM2 (« Controllable Cloning - Clone any voice from a short clip, with optional style guidance to steer emotion, pace, and expression while preserving timbre »), 17/10/2026"},{"appuie":["langues"],"lu":"https://huggingface.co/openbmb/VoxCPM2 (« Supported Languages (30) Arabic, Burmese, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Indonesian, Italian »), 17/10/2026"},{"appuie":["langues"],"lu":"https://huggingface.co/openbmb/VoxCPM2 (« Japanese, Khmer, Korean, Lao, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Tagalog, Thai, Turkish, Vietnamese »), 17/10/2026"},{"appuie":["sortie"],"lu":"https://github.com/OpenBMB/VoxCPM (« [2026.04] 🔥 We release VoxCPM2 - 2B, 30 languages, Voice Design & Controllable Voice Cloning, 48kHz audio output! Weights »), 17/10/2026"}],"classements":[{"id":"synthese/prononciation","nom":"Prononciation du français","rang":5,"sur":16,"valeur":5.46,"unite":"%","chiffres":{"rtfx":1.82,"taille_milliards":2.38,"langues":30,"clonage":true},"voisins":["supertonic-3","funaudiollm-fun-cosyvoice3-0-5b-2512"],"source":{"nom":"Open TTS Leaderboard","mention":"Source : Open TTS Leaderboard, Hugging Face","url":"https://huggingface.co/spaces/hf-audio/open_tts_leaderboard","releve":"2026-10-19"}}],"noms_dans_les_sources":{"hf_audio":"openbmb/VoxCPM2"}}