{"id":"sprag-symphony","nom":"Symphony","editeur":"Sprag","pays":"États-Unis","poids_ouverts":false,"poids_selon":"Open ASR Leaderboard","licence":null,"licence_selon":null,"liens":{"page":"https://sprag.ai/docs/stt/models","essayer":"https://sprag.ai/docs/stt"},"fiche_auto":false,"page":"https://quelleia.com/voix/modeles/sprag-symphony/","releve":"2026-10-19","en_bref":"6e sur 39 en transcription en français (ex æquo à valeur égale) ; 12e sur 53 en transcription en anglais (Open ASR).","presentation":"Symphony est un modèle multimodal de Sprag, bâti sur Qwen3-Omni, qui reçoit texte, image, audio et vidéo et répond en texte ou en voix, en temps réel. L'API de Sprag l'utilise aussi pour transcrire et traduire la parole.","langues":{"annoncees":{"annoncees":"19 langues en compréhension de la parole","nombre":19},"prises_en_charge_selon_open_tts":null,"evaluees":["fr","en"]},"preuves":[{"appuie":["presentation"],"lu":"https://sprag.ai/docs/stt/models (« Symphony symphony /v1/audio/transcriptions , /v1/audio/translations , /v1/chat/completions , or a realtime session »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://sprag.ai/models (« Realtime multimodal serve on full Qwen3-Omni (thinker + talker + code2wav): text, image, audio, and video in; text and speech out, with realtime tool calling. »), 17/10/2026"},{"appuie":["langues"],"lu":"https://sprag.ai/models (« Supports text in 119 languages, speech understanding in 19, and speech generation in 10. »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://github.com/huggingface/open_asr_leaderboard/pull/212 (« Adds a Sprag provider to the API track, serving sprag/symphony. »), 17/10/2026"}],"classements":[{"id":"transcription/francais","nom":"Transcription en français","rang":6,"sur":39,"valeur":3.92,"unite":"%","chiffres":{"jeux":{"Common Voice":5.61,"FLEURS":3.33,"MLS":2.81}},"voisins":["universal-3-pro","assemblyai-universal-3-5-pro"],"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/open-asr-anglais","nom":"Transcription en anglais, Open ASR","rang":12,"sur":53,"valeur":5.84,"unite":"%","chiffres":{"jeux":{"AMI-Cleaned":9.02,"Earnings22-Cleaned-AA-chunked":6.01,"Gigaspeech-Cleaned":7.52,"LS Clean":1,"LS Other":2.19,"Private (conversational)":12.78,"Private (scripted)":3.99,"SPGISpeech":2.21,"URGENT24":12.64,"Voice Arena Monsoon":3.55,"Voxpopuli-AA-Cleaned":3.33}},"voisins":["cohere-transcribe","hojo-asr-v1"],"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"}}],"noms_dans_les_sources":{"hf_audio":"sprag/symphony"}}