{
 "edition": "2026-W43",
 "date": "2026-10-19",
 "id": "edge0",
 "nom": "Edge0",
 "url": "https://quelleia.com/editeurs/edge0/",
 "pays": null,
 "ecritures": [
  "Edge0"
 ],
 "liens": {
  "site": "https://edge0.ai/",
  "essayer": null,
  "application": null,
  "api": null
 },
 "totaux": {
  "modeles": 3,
  "langage": 0,
  "langage_classes": 0,
  "images": 0,
  "images_classes": 0,
  "videos": 0,
  "videos_classes": 0,
  "decision": 0,
  "voix": 3
 },
 "langage": [],
 "images": [],
 "videos": [],
 "decision": [],
 "voix": [
  {
   "id": "edge0-ark-asr-3b",
   "nom": "ARK-ASR-3B",
   "url": "https://quelleia.com/voix/modeles/edge0-ark-asr-3b/",
   "classements": [
    {
     "id": "transcription/open-asr-anglais",
     "rang": 15,
     "sur": 53,
     "valeur": "5,92 %",
     "source": "Source : Open ASR Leaderboard, Hugging Face"
    }
   ]
  },
  {
   "id": "edge0-ark-asr-0-6b",
   "nom": "ARK-ASR-0.6B",
   "url": "https://quelleia.com/voix/modeles/edge0-ark-asr-0-6b/",
   "classements": [
    {
     "id": "transcription/open-asr-anglais",
     "rang": 17,
     "sur": 53,
     "valeur": "5,93 %",
     "source": "Source : Open ASR Leaderboard, Hugging Face"
    }
   ]
  },
  {
   "id": "edge0-audio8-asr-0-1b",
   "nom": "Audio8-ASR-0.1B",
   "url": "https://quelleia.com/voix/modeles/edge0-audio8-asr-0-1b/",
   "classements": [
    {
     "id": "transcription/open-asr-anglais",
     "rang": 45,
     "sur": 53,
     "valeur": "8,29 %",
     "source": "Source : Open ASR Leaderboard, Hugging Face"
    }
   ]
  }
 ],
 "presentation_voix": {
  "texte": "Edge0 conçoit des modèles de fondation (langage et voix) faits pour tourner localement sur du matériel grand public, des téléphones aux objets connectés et aux robots. Sa gamme vocale Audio8 comprend des modèles de transcription et de synthèse vocale publiés sur Hugging Face.",
  "faits": [],
  "site": "https://edge0.ai/",
  "pages_lues": [
   "https://huggingface.co/Edge0 (« Edge0 builds efficient foundation models for phones, laptops, wearables, robots, and other consumer hardware. Our models are designed for local, private, and offline inference. »), 17/10/2026",
   "https://edge0.ai/aboutus (« Building fast, compute-efficient foundation models so intelligence can be available anytime, everywhere Models Edge0-35B Edge0-8B Audio8-ASR-3B Audio8-ASR-0.6B Audio8-ASR-0.1B Audio8-TTS-0.6B Audio8-TTS-0.1B »), 17/10/2026"
  ]
 },
 "outils_qui_proposent": [],
 "outils_edites": [],
 "en_local": [],
 "voisins": [
  "shanghai-ai-laboratory",
  "elevenlabs",
  "fish-audio",
  "hugging-face"
 ]
}