{"id":"inworld-realtime-tts-2","nom":"Realtime TTS-2","editeur":"Inworld AI","pays":"États-Unis","poids_ouverts":false,"poids_selon":"Artificial Analysis","licence":null,"licence_selon":null,"liens":{"page":"https://inworld.ai/blog/realtime-tts-2","essayer":"https://platform.inworld.ai/tts"},"fiche_auto":true,"page":"https://quelleia.com/voix/modeles/inworld-realtime-tts-2/","releve":"2026-10-19","en_bref":"6e sur 31 en synthèse vocale en français, ex æquo avec 4 autres modèles dont les marges d'erreur se recouvrent ; 7e sur 96 en synthèse vocale en anglais (Artificial Analysis), dans un peloton de 5 modèles que la marge d'erreur ne permet pas de séparer.","presentation":"Modèle de synthèse vocale en temps réel d'Inworld conçu pour la conversation, qui tient compte des tours de parole précédents et se dirige par des instructions en langage naturel entre crochets. Il conserve la même voix d'une langue à l'autre, y compris au milieu d'une phrase.","sortie":"2026-05-05","langues":{"annoncees":{"annoncees":"470 langues (623 codes de langue et de région), dont le français","nombre":470,"francais":true},"prises_en_charge_selon_open_tts":null,"evaluees":["fr","en"]},"preuves":[{"appuie":["presentation","sortie"],"lu":"https://docs.inworld.ai/release-notes/tts.md (« Launched [Realtime TTS-2](https://inworld.ai/blog/realtime-tts-2) (`inworld-tts-2`), our most powerful and expressive TTS model: »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://inworld.ai/blog/realtime-tts-2 (« Realtime TTS-2 from Inworld AI is a new generation of voice model built for realtime conversation. It hears the full audio of the exchange, picks up the user's tone, pacing and emotional state »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://inworld.ai/blog/realtime-tts-2 (« One voice identity preserved across over 100 languages, including mid-utterance language switches inside a single generation. »), 17/10/2026"},{"appuie":["langues"],"lu":"https://docs.inworld.ai/tts/capabilities/multilingual.md (« The Realtime TTS-2 models accept **623 language and locale codes** (BCP-47) in the [`language`](#specifying-a-language) field, covering **470 languages** »), 17/10/2026"},{"appuie":["langues"],"lu":"https://docs.inworld.ai/tts/capabilities/multilingual.md (« | French | `fr` | `fr-FR` | »), 17/10/2026"}],"classements":[{"id":"synthese/francais","nom":"Synthèse vocale en français","rang":6,"sur":31,"valeur":1255.68,"unite":"Elo","bas":1233.68,"haut":1277.68,"votes":1122,"chiffres":{"prix_par_million_caracteres":20.83,"date_sortie":"2026-08-31"},"voisins":["elevenlabs-eleven-v3","gradium-tts-sep-2026"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/text-to-speech/model-families/inworld","releve":"2026-10-19"}},{"id":"synthese/artificial-analysis","nom":"Synthèse vocale en anglais, Artificial Analysis","rang":7,"sur":96,"valeur":1262.26,"unite":"Elo","bas":1247.26,"haut":1277.26,"votes":2011,"chiffres":{"prix_par_million_caracteres":20.83,"date_sortie":"2026-08-31"},"voisins":["alibaba-qwen-audio-3-0-tts-plus","google-gemini-3-8-flash-lite-tts"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/text-to-speech/model-families/inworld","releve":"2026-10-19"}}],"noms_dans_les_sources":{"artificial_analysis":"Realtime TTS-2"}}