{"id":"stepaudio-3-realtime","nom":"StepAudio 3 Realtime","editeur":"StepFun","pays":"Chine","poids_ouverts":false,"poids_selon":"Artificial Analysis","licence":null,"licence_selon":null,"liens":{"page":"https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-realtime","essayer":"https://audio.stepfun.ai/"},"fiche_auto":false,"page":"https://quelleia.com/voix/modeles/stepaudio-3-realtime/","releve":"2026-11-09","en_bref":"1er sur 38 en raisonnement à l'écoute (Big Bench Audio) ; 1er sur 31 en tenue de la conversation (Full Duplex Bench) ; 34e sur 34 au délai avant le premier son.","presentation":"Modèle vocal en duplex intégral de StepFun, servi par son API Realtime en préversion gratuite, qui raisonne en parlant et appelle des outils. Il perçoit aussi le ton, l'émotion et les bruits ambiants et gère les interruptions et les relances.","langues":{"annoncees":{"annoncees":"chinois et anglais","nombre":2,"francais":false,"liste":["zh","en"]},"prises_en_charge_selon_open_tts":null,"evaluees":["en"]},"preuves":[{"appuie":["presentation"],"lu":"https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-realtime.md (« Capabilities: natural human-like interaction, full-duplex dialogue, adaptive reasoning, think-while-speaking, Voice Agent »), 23/10/2026"},{"appuie":["presentation"],"lu":"https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-realtime.md (« It recognizes not only what the user says, but also tone, emotion, paralinguistic cues, and ambient sound. »), 23/10/2026"},{"appuie":["presentation"],"lu":"https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-realtime.md (« `stepaudio-3-realtime-preview` is the model name used during the free-trial period. »), 23/10/2026"},{"appuie":["langues"],"lu":"https://platform.stepfun.ai/docs/en/guides/models/audio.md (« Realtime - Chinese, English »), 23/10/2026"}],"classements":[{"id":"voix-a-voix/raisonnement","nom":"Raisonnement à l'écoute, Big Bench Audio","rang":1,"sur":38,"valeur":99.7,"unite":"%","chiffres":{"hebergeur":"StepFun"},"voisins":["alibaba-qwen-audio-3-0-realtime-plus"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/speech-to-speech","releve":"2026-11-09"}},{"id":"voix-a-voix/conversation","nom":"Tenue de la conversation, Full Duplex Bench","rang":1,"sur":31,"valeur":98.9,"unite":"%","chiffres":{"hebergeur":"StepFun"},"voisins":["alibaba-qwen-audio-3-0-realtime-plus"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/speech-to-speech","releve":"2026-11-09"}},{"id":"voix-a-voix/delai","nom":"Délai avant le premier son","rang":34,"sur":34,"valeur":8.835,"unite":"s","chiffres":{"hebergeur":"StepFun"},"voisins":["alibaba-qwen3-omni-30b-a3b-instruct-flash"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/speech-to-speech","releve":"2026-11-09"}}],"noms_dans_les_sources":{"aa_voix_a_voix":"StepAudio 3 Realtime"}}