{"id":"spacexai-grok-voice-3-0","nom":"Grok Voice 3.0","editeur":"SpaceXSI","pays":"États-Unis","poids_ouverts":false,"poids_selon":"Artificial Analysis","licence":null,"licence_selon":null,"liens":{"page":"https://x.ai/news/grok-voice-3","essayer":"https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech"},"fiche_auto":false,"page":"https://quelleia.com/voix/modeles/spacexai-grok-voice-3-0/","releve":"2026-11-09","en_bref":"6e sur 38 en raisonnement à l'écoute (Big Bench Audio) ; 4e sur 31 en tenue de la conversation (Full Duplex Bench) (ex æquo à valeur égale) ; 10e sur 34 au délai avant le premier son (ex æquo à valeur égale).","presentation":"Modèle parole à parole de SpaceXSI, servi par l'API Speech to Speech au même prix que Grok Voice 2.0, qui sert déjà dans les centres d'appels de Starlink. Il raisonne pendant qu'il parle et vise les appels de service client.","sortie":"2026-10-14","langues":{"annoncees":null,"prises_en_charge_selon_open_tts":null,"evaluees":["en"]},"preuves":[{"appuie":["presentation","sortie"],"lu":"https://x.ai/news/grok-voice-3 (« Oct 14, 2026 Introducing Grok Voice 3.0 The world’s most intelligent voice model. »), 23/10/2026"},{"appuie":["presentation"],"lu":"https://x.ai/news/grok-voice-3 (« Today, we’re announcing Grok Voice 3.0, the world’s most intelligent speech-to-speech model. It sets a new record for resolving customer support calls »), 23/10/2026"},{"appuie":["presentation"],"lu":"https://x.ai/news/grok-voice-3 (« Starlink call centers use Grok Voice to answer 20,000 customer calls every day across multiple languages. »), 23/10/2026"},{"appuie":["presentation"],"lu":"https://docs.x.ai/developers/release-notes.md (« `grok-voice-3.0` is now available with Speech to Speech, at the same price as Grok Voice 2.0. »), 23/10/2026"},{"appuie":["presentation"],"lu":"https://x.ai/news/grok-voice-think-fast-2 (« Grok Voice models have a unique characteristic: they reason through queries while speaking. »), 23/10/2026"}],"classements":[{"id":"voix-a-voix/raisonnement","nom":"Raisonnement à l'écoute, Big Bench Audio","rang":6,"sur":38,"valeur":98.5,"unite":"%","chiffres":{"cout_par_heure_audio_entree":4.8,"hebergeur":"SpaceXAI"},"voisins":["alibaba-qwen3-5-omni-plus-realtime","google-gemini-3-8-live-extended-thinking"],"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":4,"sur":31,"valeur":96.97,"unite":"%","chiffres":{"cout_par_heure_audio_entree":4.8,"hebergeur":"SpaceXAI"},"voisins":["openai-gpt-live-1","alibaba-qwen-audio-3-0-realtime-flash"],"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":10,"sur":34,"valeur":0.814,"unite":"s","chiffres":{"cout_par_heure_audio_entree":4.8,"hebergeur":"SpaceXAI"},"voisins":["openai-gpt-realtime-1-5","openai-gpt-realtime-2-1-mini"],"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":"Grok Voice 3.0"}}