{"id":"edge0-ark-asr-0-6b","nom":"ARK-ASR-0.6B","editeur":"Edge0","pays":null,"poids_ouverts":true,"poids_selon":null,"licence":"Apache 2.0","licence_selon":null,"liens":{"page":"https://edge0.ai/models/asr-0.6b"},"fiche_auto":true,"page":"https://quelleia.com/voix/modeles/edge0-ark-asr-0-6b/","releve":"2026-10-19","en_bref":"17e sur 53 en transcription en anglais (Open ASR).","presentation":"Modèle de transcription multilingue de taille moyenne, entraîné par distillation à partir d'un modèle enseignant, avec un décodeur d'environ 0,6 milliard de paramètres. Edge0 le positionne comme un compromis entre qualité, débit et coût de service, sur un seul GPU ou une station de travail.","sortie":"2026-09-10","taille":"0,6 milliard de paramètres (décodeur), plus un encodeur audio séparé d'environ 0,6 milliard","langues":{"annoncees":{"annoncees":"19 langues, dont le français","nombre":19,"francais":true,"liste":["zh","en","de","ja","fr","ko","es","pl","it","ro","hu","cs","nl","fi","hr","sk","sl","et","lt"]},"prises_en_charge_selon_open_tts":null,"evaluees":["en"]},"preuves":[{"appuie":["presentation","taille"],"lu":"https://huggingface.co/Edge0/ARK-ASR-0.6B/raw/main/README.md (« ARK-ASR-0.6B is an automatic speech recognition model trained with teacher-data adaptation and on-policy distillation, using a compact 0.6B-scale decoder LLM together with a dedicated audio encoder and adapter. »), 17/10/2026"},{"appuie":["taille"],"lu":"https://huggingface.co/Edge0/ARK-ASR-0.6B/raw/main/README.md (« 0.6B decoder LLM parameters, with a separate 0.6B-scale Whisper-style audio encoder and MLP adapter »), 17/10/2026"},{"appuie":["langues"],"lu":"https://huggingface.co/Edge0/ARK-ASR-0.6B/raw/main/README.md (« ARK-ASR currently supports Chinese, English, German, Japanese, French, Korean, Spanish, Polish, Italian, Romanian, Hungarian, Czech, Dutch, Finnish, Croatian, Slovak, Slovene, Estonian, and Lithuanian ASR. »), 17/10/2026"},{"appuie":["sortie"],"lu":"https://edge0.ai/models/asr-0.6b (« Audio8 · ASR · 0.6B September 10, 2026 »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://edge0.ai/models/asr-0.6b (« Audio8 ASR 0.6B balances transcription quality, throughput, memory use and serving cost for practical production deployment. »), 17/10/2026"}],"classements":[{"id":"transcription/open-asr-anglais","nom":"Transcription en anglais, Open ASR","rang":17,"sur":53,"valeur":5.93,"unite":"%","chiffres":{"rtfx":660.66,"jeux":{"AMI-Cleaned":8.61,"Earnings22-Cleaned-AA-chunked":6.75,"Gigaspeech-Cleaned":7.37,"LS Clean":1.48,"LS Other":3.44,"Private (conversational)":12.57,"Private (scripted)":3.68,"SPGISpeech":2.63,"URGENT24":12.49,"Voice Arena Monsoon":3.43,"Voxpopuli-AA-Cleaned":2.76}},"voisins":["boson-ai-higgs-audio-v3-stt","canary-qwen-2-5b"],"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":"Edge0/ARK-ASR-0.6B"}}