agreed. challenge is if/when they start optimizing for this revenue stream
INNOVATION
-
Hume AI Voice Cloning Tool Enables Realistic Self-Replication
By
–
i just tested @hume_ai's upcoming voice cloning tool
— Yohei (@yoheinakajima) 16 juillet 2025
uploaded a ~40 sec audio of myself reading a blog posts to test, and it sounds like i'm talking to myself! pic.twitter.com/99FfR6HaAci just tested @hume_ai
's upcoming voice cloning tool uploaded a ~40 sec audio of myself reading a blog posts to test, and it sounds like i'm talking to myself! -
Voice AI Agent Clones Speaking Style and Personality
By
–
seems like it didn't just clone my voice, but my speaking style, and even used the content of the audio to define the "personality" of the voice agent
-
From $500K AI Course Success to Single Sale Lesson
By
–
He made $500K selling AI courses.
— AppSumo.com (@AppSumo) 16 juillet 2025
Then his next course sold… one seat.
Here’s what @thatroblennon learned about this experience: pic.twitter.com/ERvDjt2FrJHe made $500K selling AI courses. Then his next course sold… one seat. Here’s what @thatroblennon learned about this experience:
→ View original post on X — @thatroblennon, 2025-07-16 22:00 UTC
-

AI Video Editing Tool Shows Imperfect Results in Demo
By
–
¿Es perfecto? No. En algunos casos cuando la edición es más compleja os daréis cuenta de que más detalles empezarán a cambiar. En el propio vídeo de demo que han enseñado reconocen que la similitud no siempre se mantiene al 100%, como podéis ver aquí.
-
Document Embeddings: Arithmetic and Text Decoding Techniques
By
–
I built this during my PhD! you can embed *documents* using the OpenAI embeddings, do arithmetic on them, and then decode them back to text e.g. E = emb(“Elon Musk bought Twitter in 2022”) – emb(“Elon Musk”)
vec2text(e) = “Someone bought twitter in 2022” -
Diffusion Infrastructure Will Outperform Autoregressive Models
By
–
one thing that in retrospect is obvious but i never considered properly is that infra for diffusion is going to outperform infra for autoregression because 1) more opportunity to optimize, 2) better open models, 3) less competition
-

Mistral Launches Audio Transcription API Endpoint
By
–
The good news is that Mistral also offer a new dedicated /v1/audio/transcriptions endpoint which appears to transcribe audio without being confused by instructions embedded in that audio
-

Voxtral Models Integration with Audio File Support
By
–
These new Voxtral models look really useful! I got them working in my llm-mistral plugin, which can now accept URLs to audio files as attachments to pass to voxtral-small and voxtral-mini
-
Hugging Face Releases Desktop Robot for AI Experimentation and Pre-Order
By
–
Hugging Face Makes Desktop #Robot Available “as-is” for Pre-Order (with no warranties), for #AI experimentation: https://
iotworldtoday.com/robotics/huggi
ng-face-makes-desktop-robot-available-for-pre-order
… What use cases would you consider for this desktop robot?