{"id":"soniox-v5-async","nom":"Soniox v5 Async","editeur":"Soniox","pays":"États-Unis","poids_ouverts":false,"poids_selon":"Artificial Analysis","licence":null,"licence_selon":null,"liens":{"page":"https://soniox.com/docs/stt/models","essayer":"https://soniox.com/docs/stt/get-started"},"fiche_auto":false,"page":"https://quelleia.com/voix/modeles/soniox-v5-async/","releve":"2026-10-19","en_bref":"23e sur 39 en transcription en français ; 24e sur 51 en transcription en anglais (Artificial Analysis).","presentation":"stt-async-v5 est le modèle de transcription de Soniox pour les fichiers audio enregistrés, servi par API. Il apporte une séparation des locuteurs refaite, une meilleure identification des langues et une transcription plus fiable des nombres, dates et adresses.","sortie":"2026-06-11","langues":{"annoncees":{"annoncees":"plus de 60 langues, dont le français","nombre":60,"plus":true,"francais":true},"prises_en_charge_selon_open_tts":null,"evaluees":["fr","en"]},"preuves":[{"appuie":["sortie"],"lu":"https://soniox.com/docs/stt/models (« June 11, 2026 New models: stt-async-v5 Replaces: stt-async-v4 »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://soniox.com/docs/stt/models (« stt-async-v5 is the new Soniox async speech-to-text model for processing recorded audio »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://soniox.com/docs/stt/models (« Completely reengineered speaker separation for identifying who said what »), 17/10/2026"},{"appuie":["langues"],"lu":"https://soniox.com/docs/stt/models (« Higher transcription accuracy across 60+ languages »), 17/10/2026"},{"appuie":["langues"],"lu":"https://soniox.com/docs/stt/concepts/supported-languages.mdx (« | French | fr | »), 17/10/2026"}],"classements":[{"id":"transcription/francais","nom":"Transcription en français","rang":23,"sur":39,"valeur":6.21,"unite":"%","chiffres":{"jeux":{"Common Voice":9.84,"FLEURS":4.24,"MLS":4.55}},"voisins":["voxtral-mini","whisper-large-v3"],"source":{"nom":"Open ASR Leaderboard","mention":"Source : Open ASR Leaderboard, Hugging Face","url":"https://huggingface.co/spaces/hf-audio/open_asr_leaderboard","releve":"2026-10-19"}},{"id":"transcription/artificial-analysis","nom":"Transcription en anglais, indice AA-WER","rang":24,"sur":51,"valeur":3.81,"unite":"%","chiffres":{"prix_pour_1000_minutes":1.66,"jeux":{"AA-AgentTalk":3.11,"Earnings22":6.7,"VoxPopuli":2.31},"date_sortie":"2026-06-11","hebergeur":"Soniox"},"voisins":["voxtral-mini-transcribe-2","assemblyai-universal"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/speech-to-text/models/soniox","releve":"2026-10-19"}}],"noms_dans_les_sources":{"hf_audio":"soniox/stt-async-v5","artificial_analysis":"Soniox v5 Async"}}