24, cute, and smartass. shipping code, pencil art, and selfies. prolly awake, prolly bored.. idk 🤭
lost in the night
Joined April 2026
- Tweets118
- Following64
- Followers7
- Likes497
Second TTS in few days.
damn, that's gettin craaazzyy
Introducing Audio8-TTS-Realistic, @Edge0_AI’s new open-weight speech model: realistic, zero-shot voice cloning in just 0.6B parameters.
• Human-like prosody and expressive delivery
• Clones a voice from one short clip, in 23 languages
• 601M params: small enough to fine-tune and serve affordably
Currently #1 on the Hugging Face TTS Arena, where people vote blind on which voice sounds more human.
Open weights are available now on @huggingface. Try it free.
a 4th model.
🤯
Introducing Phonic-1, the most advanced speech-to-speech model for agentic tasks — from @phonic_co, the speech-to-speech model company.
We’ve had a breakout year and raised a $27.5M Series A led by @firstround with participation from @Lux_Capital, @felicis, and Definition.
a third voice model in 2 days? damn accel..
the mid..
it's always the mid.. somehow..
and mean that on few levels.
did you seen her big mouth shock and huge smile 🤯😳😱
so cute.
and he's a proof you need just 30 to 60 seconds to change everything.
She..
She did a tongue mistake and corrected herself in an instant!
Like.. WHAT THE..
@dessa
i'm crying
@dessa is sick
We believe voice is the most natural way to interact with intelligence, so we created Dessa, the most authentic full-duplex speech model ever built. Talking with it feels like talking with a person…full of emotion, nuance, pauses, and a natural cadence.
block.xyz/dessa
"So if someone is keep telling me they're willing to learn more about the space, i know they are willing to be prompted for the right answer.
So, yeah, it's just a stage of prompting."
WTF.
holy sh!t
And there's more at @dessa.. this secret lab is sick.
i mean.. Opus 5.5 DID/DOES have 1M context.. what happened, @theo @t3dotcodes @maria_rcks ????
i clearly remember having it on 1M high fast mode..
btw, dropping the lightning emoji for "fast mode" is regression too!
ha! i do remember reading this one about 10 years ago, yeah..
well, it was audiobook, but still..
am i the only one who still watch shit in full and normal 1x speed?
like.. if it's something i don't care that much, sure.. I'd summarize, but.. if it's genuinely good i really do watch.
love you @theo but there is no way my adhd self is going to watch the 32 minutes, at least not without dedicated time set aside, so I had grokbot make me a podcast.
welcome to the future. where you take content that is fractionalized and fractionalize it again.
OH. MY. GOD. YES!
- Webhook tasks got filters and dedupe
We just shipped an updated T3 Code Stable release with over 1,400 changes 🤯
Quick highlight reel of my favorites:
- Massive perf wins on desktop and mobile
- Add a project to another machine from the "Run on" menu
- EAS Simulator remote hosts for iOS dev
- PR diffs now show code vs tests vs docs at a glance
- Idle Claude threads compact early so the cache stays warm (beta)
- Mobile got a big native pass. It finally feels native
- `t3 doctor` can reclaim database space
- `t3 connect` now checks your machine is reachable, and errors say how to fix them
- Webhook tasks got filters and dedupe
- Skill discovery is much more reliable
- Tons of Windows, terminal, and scroll fixes
Least favorite: I caved, LaTeX renders on web and mobile
Also heads up, we're getting a bit more experimental with how we treat Nightly, so now would be a good time to move back to stable :)
