{"id":"voxtral-tts","nom":"Voxtral TTS","editeur":"Mistral AI","pays":"France","poids_ouverts":true,"poids_selon":null,"licence":"CC BY-NC 4.0","licence_selon":null,"liens":{"poids":"https://huggingface.co/mistralai/Voxtral-4B-TTS-2603","page":"https://mistral.ai/news/voxtral-tts","essayer":"https://docs.mistral.ai/studio/audio/text_to_speech"},"fiche_auto":false,"page":"https://quelleia.com/voix/modeles/voxtral-tts/","releve":"2026-10-19","en_bref":"23e sur 31 en synthèse vocale en français, ex æquo avec 3 autres modèles dont les marges d'erreur se recouvrent ; 14e sur 16 en prononciation du français ; 45e sur 96 en synthèse vocale en anglais (Artificial Analysis), dans un groupe de 10 modèles trop proches pour être départagés.","presentation":"Premier modèle de synthèse vocale de Mistral AI, avec des poids publiés sous licence non commerciale, 20 voix prédéfinies et l'imitation d'une voix à partir d'un court extrait sans transcription. Mistral annonce environ 90 millisecondes avant le premier son en diffusion continue.","sortie":"2026-03-23","taille":"4 milliards de paramètres","langues":{"annoncees":{"annoncees":"9 langues, dont le français","nombre":9,"francais":true,"liste":["en","fr","es","de","it","pt","nl","ar","hi"]},"prises_en_charge_selon_open_tts":9,"evaluees":["fr","en"]},"preuves":[{"appuie":["sortie"],"lu":"https://mistral.ai/news/voxtral-tts (« Speaking of Voxtral March 23, 2026 By Mistral »), 17/10/2026"},{"appuie":["presentation","taille"],"lu":"https://mistral.ai/news/voxtral-tts (« Today we’re releasing Voxtral TTS, our first text-to-speech model with state-of-the-art performance in multilingual voice generation. The model is lightweight at 4B parameters »), 17/10/2026"},{"appuie":["langues","presentation"],"lu":"https://docs.mistral.ai/models/voxtral-tts-26-03 (« Our state-of-the-art text-to-speech model with zero-shot voice cloning. Supports 9 languages, streaming with ~90ms time-to-first-audio, and no transcript required for voice prompts. »), 17/10/2026"},{"appuie":["langues"],"lu":"https://huggingface.co/mistralai/Voxtral-4B-TTS-2603/raw/main/README.md (« - **Multilingual support**: English, French, Spanish, German, Italian, Portuguese, Dutch, Arabic, and Hindi »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/mistralai/Voxtral-4B-TTS-2603/raw/main/README.md (« - **Text-to-Speech generation** with 20 preset voices and easy adaptation to new voices »), 17/10/2026"}],"classements":[{"id":"synthese/francais","nom":"Synthèse vocale en français","rang":23,"sur":31,"valeur":1063.73,"unite":"Elo","bas":1042.73,"haut":1084.73,"votes":1179,"chiffres":{"prix_par_million_caracteres":16,"date_sortie":"2026-03-26"},"voisins":["soniox-tts-real-time-v2","fish-audio-openaudio-s1-mini"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/text-to-speech/model-families/mistral","releve":"2026-10-19"}},{"id":"synthese/prononciation","nom":"Prononciation du français","rang":14,"sur":16,"valeur":13.42,"unite":"%","chiffres":{"rtfx":50.46,"taille_milliards":4,"langues":9,"clonage":false},"voisins":["microsoft-vibevoice-realtime-0-5b","alibaba-qwen3-omni-30b-a3b-instruct"],"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"}},{"id":"synthese/artificial-analysis","nom":"Synthèse vocale en anglais, Artificial Analysis","rang":45,"sur":96,"valeur":1089.21,"unite":"Elo","bas":1077.21,"haut":1101.21,"votes":2345,"chiffres":{"prix_par_million_caracteres":16,"date_sortie":"2026-03-26"},"voisins":["openai-tts-1","google-gemini-2-5-flash-lite-tts"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/text-to-speech/model-families/mistral","releve":"2026-10-19"}}],"noms_dans_les_sources":{"artificial_analysis":"Voxtral TTS","hf_audio":"mistralai/Voxtral-4B-TTS-2603"}}