{"id":"microsoft-vibevoice-realtime-0-5b","nom":"VibeVoice-Realtime-0.5B","editeur":"Microsoft","pays":"États-Unis","poids_ouverts":true,"poids_selon":null,"licence":"MIT","licence_selon":null,"liens":{"page":"https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B"},"fiche_auto":true,"page":"https://quelleia.com/voix/modeles/microsoft-vibevoice-realtime-0-5b/","releve":"2026-10-19","en_bref":"13e sur 16 en prononciation du français.","presentation":"Modèle ouvert de synthèse vocale en temps réel de Microsoft, qui accepte du texte en flux et produit un premier son en environ 300 ms. Il ne gère qu'un seul locuteur et vise surtout l'anglais.","sortie":"2025-12-03","taille":"0,5 milliard de paramètres","langues":{"annoncees":{"annoncees":"anglais, plus neuf langues expérimentales dont le français","francais":true,"liste":["en","de","fr","it","ja","ko","nl","pl","pt","es"]},"prises_en_charge_selon_open_tts":10,"evaluees":["fr"]},"preuves":[{"appuie":["presentation"],"lu":"https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B/raw/main/README.md (« VibeVoice-Realtime is a **lightweight real‑time** text-to-speech model supporting **streaming text input** and **robust long-form speech generation**. »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B/raw/main/README.md (« - Realtime TTS (~300 ms first audible latency) »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B/raw/main/README.md (« This realtime variant supports only a single speaker. »), 17/10/2026"},{"appuie":["langues"],"lu":"https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B/raw/main/README.md (« We provide nine additional languages (German, French, Italian, Japanese, Korean, Dutch, Polish, Portuguese, and Spanish) for users to explore and share feedback. »), 17/10/2026"},{"appuie":["langues"],"lu":"https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B/raw/main/README.md (« The model is currently intended for English speech only; other languages may produce unpredictable results. »), 17/10/2026"},{"appuie":["taille"],"lu":"https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B/raw/main/README.md (« - Parameter size: 0.5B (deployment-friendly) »), 17/10/2026"},{"appuie":["sortie"],"lu":"https://raw.githubusercontent.com/microsoft/VibeVoice/main/README.md (« 2025-12-03: 📣 We open-sourced »), 17/10/2026"}],"classements":[{"id":"synthese/prononciation","nom":"Prononciation du français","rang":13,"sur":16,"valeur":11.92,"unite":"%","chiffres":{"rtfx":2.52,"taille_milliards":1.36,"langues":10,"clonage":false},"voisins":["boson-ai-higgs-tts-3-4b","voxtral-tts"],"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":"microsoft/VibeVoice-Realtime-0.5B"}}