Mega-TTS: Zero-Shot Text-to-Speech at Scale with Intrinsic Inductive Bias paper page: https://
huggingface.co/papers/2306.03
509
… Scaling text-to-speech to a large and wild dataset has been proven to be highly effective in achieving timbre and speech style generalization, particularly in zero-shot
MULTIMODAL AI
-
Mega-TTS: Zero-Shot Text-to-Speech Scaling with Inductive Bias
By
–
-
Generative AI Explained: Visual Guide to Next Big Thing
By
–
generative artificial intelligence could be the next big thing, so check out this great infographic Infographic: Generative AI Explained by AI https://
buff.ly/3DzhuNo
#ai #ArtificialIntelligence #MachineLearning #deeplearning #GenerativeAI #infographic #dalle #MidjourneyAI #art -
AI Visual Captions System Enhances Online Meeting Communication
By
–
What if #AI could show context-relevant visuals in online meetings? #VisualCaptions in #ARChat is an interactive system to augment communication w/ interactive visuals suggested by #LLM + a search →https://t.co/NZg0qVemYt
— Google AI (@GoogleAI) 6 juin 2023
Fork & star our code & dataset →https://t.co/TFWYx8sjzf pic.twitter.com/LpfSTT9hJgWhat if #AI could show context-relevant visuals in online meetings? #VisualCaptions in #ARChat is an interactive system to augment communication w/ interactive visuals suggested by #LLM + a search →
https://
goo.gle/42yYgBi
Fork & star our code & dataset →
https://
github.com/google/archat -
Video-LLaMA Paper Lacks Quantitative Evaluation Metrics
By
–
"Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Understanding". This method is pretty cool, but what is this new trend where papers don't contain any type of quantitative model evaluation? Looks like researchers are in a rush. https://
arxiv.org/abs/2306.02858 -
Local LLMs reduce latency for real-time NPC conversations
By
–
So one problem with using GPT-3 or 4 in games for NPC conversations is latency. Best case you can bring it down to 2s b/w input and output. But the onboard M2 on the Vision can allow you to use local LLMs which can bring latencies down to allow for things like realtime therapy
-
Video AI Struggles with Frame Consistency and Narrative
By
–
frame to frame consistency is poor. Plus if you actually try putting narrative into it: it fails. text2img is easier because narrative isn't required.
-
Neuralangelo: High-Fidelity Neural Surface Reconstruction Method
By
–
Neuralangelo: High-Fidelity Neural Surface Reconstruction
— AK (@_akhaliq) 6 juin 2023
paper page: https://t.co/omcwjlyeha
present Neuralangelo, which combines the representation power of multi-resolution 3D hash grids with neural surface rendering. Two key ingredients enable our approach: (1) numerical… pic.twitter.com/n3beH5Oa4ENeuralangelo: High-Fidelity Neural Surface Reconstruction paper page: https://
huggingface.co/papers/2306.03
092
… present Neuralangelo, which combines the representation power of multi-resolution 3D hash grids with neural surface rendering. Two key ingredients enable our approach: (1) numerical -
GPT Models Meet Robotic Applications: Co-Speech Gesturing Chat System
By
–
GPT Models Meet Robotic Applications: Co-Speech Gesturing Chat System paper page: https://
huggingface.co/papers/2306.01
741
… introduces a chatting robot system that utilizes recent advancements in large-scale language models (LLMs) such as GPT-3 and ChatGPT. The system is integrated with a -
HeadSculpt: Text-Guided 3D Head Avatar Generation Method
By
–
HeadSculpt: Crafting 3D Head Avatars with Text
— AK (@_akhaliq) 6 juin 2023
paper page: https://t.co/DeebCTOHxp
Recently, text-guided 3D generative methods have made remarkable advancements in producing high-quality textures and geometry, capitalizing on the proliferation of large vision-language and image… pic.twitter.com/JOGeVMvgbIHeadSculpt: Crafting 3D Head Avatars with Text paper page: https://
huggingface.co/papers/2306.03
038
… Recently, text-guided 3D generative methods have made remarkable advancements in producing high-quality textures and geometry, capitalizing on the proliferation of large vision-language and image