Agreed @tdietterich that touch is part of grounding. It’s a very precise and high fidelity grounding in a highly localized way (typically within square centimeters or at best square feet scale), whereas vision is a much more expansive grounding of the world that can extend very
MULTIMODAL AI
-
CapSpeech: Guidelines for Expressive Text-to-Speech Systems
By
–
CapSpeech provides guidelines for data construction & model design to develop prompt-guided expressive text-to-speech systems across use cases like cinematic narration, game character voiceovers, conversational chatbots, emotion therapy, & more.
-
CapSpeech: AI Text-to-Speech with Controllable Voice Design
By
–
MIT CSAIL, JHU, & USC researchers have developed "CapSpeech," a text-to-speech framework that generates voices w/controllable timbre & speaking style via text prompts.
— MIT CSAIL (@MIT_CSAIL) 5 juin 2025
Imagine designing a voice just by describing it. CapSpeech can customize age, timbre, accent, emotion, & more:… pic.twitter.com/Dw0i6JUG19MIT CSAIL, JHU, & USC researchers have developed "CapSpeech," a text-to-speech framework that generates voices w/controllable timbre & speaking style via text prompts. Imagine designing a voice just by describing it. CapSpeech can customize age, timbre, accent, emotion, & more:
-
Face-to-Many Flux Model Released on Replicate
By
–
I made a thing https://
replicate.com/flux-kontext-a
pps/face-to-many-kontext
… -
Voice Mode Update Coming Soon for AI Assistant
By
–
Yep, tiene pinta de que pronto veremos un update del modo de voz! Se celebra 🍺https://t.co/DdFK8eoaaX
— Carlos Santana (@DotCSV) 5 juin 2025Yep, tiene pinta de que pronto veremos un update del modo de voz! Se celebra
-
Potential Impact of Sora Integration into Microsoft Bing
By
–
This is the real unlock, zero onboarding, mobile-first, no cost. Sora inside Bing might be Microsoft’s next $100B ecosystem play.
-
ChatGPT Voice Mode: Useful But Needs Polish
By
–
El modo de voz de ChatGPT es de esas cosas que está en la frontera de ser súper útil si terminan de pulir ciertos detalles, como la naturalidad de la voz o la profundidad de las conversaciones. Durante este año ha estado rozando la frontera, pero al menos para mi caso sin pasar
-
ChatGPT Voice Mode Improves Accent Recognition Across Languages
By
–
También esta semana me contaban el caso en el que de un día para otro el modo de voz de ChatGPT ya no respondía con acento argentino cuando le hablabas en canario, sino que ya su acento se ajustaba mucho mejor.
— Carlos Santana (@DotCSV) 5 juin 2025
Tiene pinta de que el AVM se está actualizando! 🙂 https://t.co/Z8S8ANC5yUTambién esta semana me contaban el caso en el que de un día para otro el modo de voz de ChatGPT ya no respondía con acento argentino cuando le hablabas en canario, sino que ya su acento se ajustaba mucho mejor. Tiene pinta de que el AVM se está actualizando! 🙂
-
Luma Labs Launches Modify Video, Outperforms Runway V2V
By
–
Luma Labs dropped Modify Video, a tool to restyle videos by changing style, settings, and more
— Rowan Cheung (@rowancheung) 5 juin 2025
In blind tests, it outperformed Runway V2V across key criteria, retaining motion, facial animation, and temporal consistency
Available in Dream Machine: Ray 2pic.twitter.com/C60XzUDUNzLuma Labs dropped Modify Video, a tool to restyle videos by changing style, settings, and more In blind tests, it outperformed Runway V2V across key criteria, retaining motion, facial animation, and temporal consistency Available in Dream Machine: Ray 2
-
EVA AI Creates Photorealistic 3D Avatars from Simple Video
By
–
Créer un avatar 3D photoréaliste d’une personne à partir d’une simple vidéo. Avec mouvements, expressions, corps complet.
— VISION IA (@vision_ia) 5 juin 2025
C’est ce que propose EVA — l’un des projets IA les plus bluffants pour le métavers.
Mais on sait déjà ce que ça permet. Et ce que ça annonce. https://t.co/c0UiRSSYrI pic.twitter.com/5kl4fVyhFkCréer un avatar 3D photoréaliste d’une personne à partir d’une simple vidéo. Avec mouvements, expressions, corps complet. C’est ce que propose EVA — l’un des projets IA les plus bluffants pour le métavers. Mais on sait déjà ce que ça permet. Et ce que ça annonce.