Unterschiede

Hier werden die Unterschiede zwischen zwei Versionen angezeigt.

Link zu dieser Vergleichsansicht

Beide Seiten der vorigen RevisionVorhergehende Überarbeitung
Nächste Überarbeitung		Vorhergehende Überarbeitung
ai_ki [2025/10/29 14:06] – [Im Web] muonitai_ki [2026/10/13 11:20] (aktuell) – [Bilder/Video] muonit
Zeile 1: Zeile 1:
 ====== AI Kram ====== ====== AI Kram ======
 +=====Performance =====
 +- https://github.com/ggml-org/llama.cpp/discussions/4167 M-Prozessoren
 +
 +
 ===== Chatbots ===== ===== Chatbots =====
 +  * https://chatjimmy.ai/ > 15000 Token/s - allerdings bei nur 8 Mrd. Parametern
   * https://playground.allenai.org/   * https://playground.allenai.org/
   * https://chat.mistral.ai/chat   * https://chat.mistral.ai/chat
Zeile 19: Zeile 24:
   * https://handy.computer/   * https://handy.computer/
   * https://github.com/cydanix/whisperclip Mit lokalem LLM zur Verbesserung danach   * https://github.com/cydanix/whisperclip Mit lokalem LLM zur Verbesserung danach
 +  * https://github.com/ufal/SimulStreaming SimulStreaming implements Whisper model for translation and transcription in simultaneous mode (which is known as streaming in the ASR community)
 +  * https://github.com/moonshine-ai/moonshine?tab=readme-ov-file
  
  
Zeile 35: Zeile 42:
   * https://x-minus.pro/ai Vocals aus jedem Song entfernen   * https://x-minus.pro/ai Vocals aus jedem Song entfernen
   * https://www.picpik.ai   * https://www.picpik.ai
 +  * https://www.latentsync.org/de Latent Sync nutzt die Leistung von Stable Diffusion und TREPA, um präzise, hochauflösende Lippensynchronisation für dynamische und realistische Videogenerierung zu liefern.
 +  * https://www.wav2lip.org/de Erstellen Sie lippen-synchronisierte Videos aus beliebigem Audio mit Wav2Lip. Probieren Sie unser kostenloses Online-Tool aus, um sofort präzise, hochqualitative Sprechgesicht-Animationen zu erzeugen.
 +  * https://muapi.ai/open-generative-ai
  
  
Zeile 43: Zeile 53:
   * https://diffusionbee.com toolbox to run AI apps locally with Stable Diffusion   * https://diffusionbee.com toolbox to run AI apps locally with Stable Diffusion
   * https://github.com/llama-farm/llamafarm    * https://github.com/llama-farm/llamafarm 
 +  * https://news.ycombinator.com/item?id=45771870 Ask HN: Who uses open LLMs and coding assistants locally? Share setup and laptop
 +  * https://github.com/ssrajadh/sentrysearch Semantic search over videos using Gemini Embedding 2 or Qwen3-VL. 
 +  * https://github.com/mahimairaja/voiceai learning path for building real-time voice AI agents
 ==== Chat ==== ==== Chat ====
   * https://github.com/mezbaul-h/june Local voice chatbot for engaging conversations, powered by Ollama, Hugging Face Transformers, and Coqui TTS Toolkit    * https://github.com/mezbaul-h/june Local voice chatbot for engaging conversations, powered by Ollama, Hugging Face Transformers, and Coqui TTS Toolkit 
Zeile 51: Zeile 64:
 ==== Bilder/Video ==== ==== Bilder/Video ====
 === Hintergrund entfernen === === Hintergrund entfernen ===
 +  * https://usefeyn.com/multimatte/
   * https://bannerify.co/tools/remove-bg   * https://bannerify.co/tools/remove-bg
   * https://github.com/pinokiofactory/RMBG-2-Studio?tab=readme-ov-file   * https://github.com/pinokiofactory/RMBG-2-Studio?tab=readme-ov-file
   * https://github.com/azizaydi23/video-bg-remover   * https://github.com/azizaydi23/video-bg-remover
 +  * https://cleanup.pictures/ Remove unwanted objects from image
 +  * https://github.com/danielgatis/rembg Rembg is a tool to remove images background 
 +  * https://github.com/feyninc/nobg a library for background removal models
  
  
Zeile 62: Zeile 79:
   * https://github.com/deepbeepmeep/Hunyuan3D-2GP 3D Model from Image   * https://github.com/deepbeepmeep/Hunyuan3D-2GP 3D Model from Image
   * https://seedream.run    * https://seedream.run 
 +  * https://github.com/apple/ml-sharp Sharp Monocular View Synthesis in Less Than a Second 
 +  * https://github.com/ArcticLatent/Arctic-Helper A Linux and Windows helper app to install, update, and manage ComfyUI, including easy downloads of AI models and LoRAs. 
 +  * https://prismml.com/news/bonsai-image-4b Bonsai Image 4B: Image Generation for Local Devices
 +  * https://z-image-turbo.com/de Erstelle Produktfotos, Porträts und Poster aus einer Beschreibung. Lade Bilder zum Bearbeiten oder Verbessern hoch.
  
 === Clone Movement to Avatar/Videogenerierung === === Clone Movement to Avatar/Videogenerierung ===
Zeile 75: Zeile 96:
   * https://github.com/yl4579/StyleTTS2   * https://github.com/yl4579/StyleTTS2
   * https://docs.hyprnote.com/owhisper/what-is-this OWhisper SST für Hyprnote   * https://docs.hyprnote.com/owhisper/what-is-this OWhisper SST für Hyprnote
 +  * https://www.hume.ai/blog/opensource-tada Fast, Reliable Speech Generation Through Text-Acoustic Synchronization
  
 ==== STT ==== ==== STT ====