Whisper uses a transformer architecture like GPT-4 and is trained on over 600,000 hours of varied web content, achieving near-human understanding of real-world audio.
MULTIMODAL AI
-
Creator launches custom GPT for image recreation
By
–
Built a Recreate Images GPT, just upload your photo and it tries to recreate it immediately. Try it out here and let me know what you think:
-
New CustomGPT recreates images from uploaded photos
By
–
ChatGPT was just the beginning. Now everyone's building Custom GPTs. My newest CustomGPT is Recreate Images GPT. Just upload your photo and watch its magic Try it out here: https://
chat.openai.com/g/g-DqViiXYrU-
recreate-images-gpt
… #RecreateImages #CustomGPT #GPTs #GPTStore #GPT4 #GPTBuilder #GPT4Vision -
Text-to-Speech Limitations and Unexpected Behavior Exploration
By
–
I’ve thrown a lot of weird prompts at text-to-speech and as yet haven’t got it to do anything unexpected. You can get it to [beep] its own swears, and occasionally an interjection might half work. Not as wild as something like Bark though.
-
Morning Tranquility Created with Gen-2 AI
By
–
Morning tranquility. Made with Gen-2. pic.twitter.com/f8i7bOyFWY
— Runway (@runwayml) 15 novembre 2023Morning tranquility. Made with Gen-2.
-
Taylor Swift as Toy Story Character Using AI Image Generation
By
–
Some examples:
Taylor Swift as a Toy Story person – toy story lora
– img2img
– depth controlnet
– pose controlnet https://
replicate.com/p/fr4k2p3bxjbr
mwsnosxtjczxcm
… -
SDXL Multi-ControlNet Model with LoRA Support Released
By
–
I've put together a new Replicate SDXL model, it does a lot: – up to 3 controlnets (+ strength, start, end controls)
– controlnet with inpainting/img2img
– canny, QR monster, pose, soft edge, depth, line art
– lora loading https://
replicate.com/fofr/sdxl-mult
i-controlnet-lora
… Please give it a spin -
Qwen-Audio: Universal Audio Understanding via Large-Scale Models
By
–
Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models Chu et al.: https://
arxiv.org/abs/2311.07919 #ArtificialIntelligence #DeepLearning #MachineLearning -
DALL-E 3 Realism GPT for Photorealistic Images
By
–
DALL-E 3 is able to create realistic images, or at least it tries. If you need photorealistic images, try out my Realism GPT here: https://
chat.openai.com/g/g-rQbzHQkZ3-
realism-gpt
… My custom Realism GPT is built for maximizing realistic photography in every output. #GPTs #GPTBuilder #GPTStore #ChatGPT4