
Software Engineer – @crosbylegal | @ITP_NYU 2022
Brooklyn, NY
Joined June 2014
- Tweets2.2K
- Following1.3K
- Followers389
- Likes12.2K
Vince Picone retweeted
Replying to @tryramp
@tryramp is growing faster this month (as a 60B company) than it was 12 months ago.
Doing a fireside chat with @rahulgs Ramp’s CTO to talk about how engineering makes that possible:
- How do you keep code review from becoming the bottleneck?
- What makes an exceptional engineer now, and how do you interview for it?
- Should product, engineering, or domain experts own evals?
Only 10 slots remaining
Wednesday, 10/21 at 6 pm. Technical audience only. DM or comment to RSVP.
Vince Picone retweeted
Some more innovation in duck based transport. Maybe they can all seperate and go and buy things from different stores then come back.
Vince Picone retweeted
The people getting great UI results with AI aren’t using a secret model or a magic prompt. They know what a great interface looks like, and they know how to steer AI towards it.
Work that used to take a week takes a day, and the quality bar doesn’t drop, it goes up, because you spend your time on the things that matter.
At the end of the day, people that use AI best are the ones that were excellent before LLMs as well. The ones that got the fundamentals right, went that extra mile, and cared about the thing they were working on.
Today we're releasing our first open source TTS model, TADA!
TADA (Text Audio Dual Alignment) is a speech-language model that generates text and audio in one synchronized stream to reduce token-level hallucinations and improve latency.
This means:
→ Zero content hallucinations across 1,000+ test samples
→ 5x faster than similar-grade LLM-based TTS
→ Fits much longer audio: 2,048 tokens cover ~700 seconds with TADA vs. ~70 seconds in conventional systems
→ Free transcript alongside audio with no added latency
Vince Picone retweeted
the terminal is quietly becoming one of the most interesting surfaces in interaction design
i've been experimenting with what a personal website feels like when it lives in the terminal instead of a browser
Today, we're announcing significant new traction in our voice AI research and development partnerships.
Over the past few months, we’ve quietly been expanding our focus on providing frontier voice AI training data to research labs looking to imbue emotion understanding and cutting edge expressivity into their foundation models.
We believe some of the most exciting possibilities for deeper speech and emotion understanding will come to fruition this year. By training today’s frontier models to understand the nuances voice interaction—rife with subtle tones of frustration or satisfaction, “aha” moments, chuckles, sighs, backchannels, and interruptions—we believe that labs will unlock new possibilities for voice AI to become a primary interface within many applications.
In our experience working with leading research labs and AI-first enterprises, a consistent pattern we’ve seen is a greater need for high-quality datasets and evaluation pipelines than for new algorithms or architectures. Researchers spend up to 80% of their time curating the data they need to diagnose model issues, fix failure modes, and improve model behavior.
That’s why Hume is now focused on building the data and evaluation infrastructure needed to train next-generation voice models across industry. With the right training, we hope that deeper voice understanding and empathy can be translated not just into more efficient interfaces but into better alignment of AI with human well-being.
Read more below hume.ai/blog/data-blog-jan
Vince Picone retweeted
Today we honor Dr. Paul Ekman, whose pioneering work on how we express and perceive emotion laid much of the groundwork our field stands on today.
His ideas still shape how we try to understand human feeling with nuance and respect. May he rest in peace.
paulekman.com/blog/the-passi…
Vince Picone retweeted
One performance, infinite voices.
Voice Conversion is now live on Hume’s creator studio and API!
Generate the same pacing, pronunciation, and intonation with one recording across any voice you choose.
Hear it for yourself ⬇️
Vince Picone retweeted
Why are phone numbers so hard for voice AI?📞🤖
Did you know that reading out something as simple as 555-010-2314 is a weirdly difficult task for TTS?
In a recent eval we ran on all the major TTS models, Hume’s Octave 2 comes out at the top with the lowest error rates for phone numbers (without sacrificing its <100ms model latency, high expressivity, and multilingual capabilities).
Why is a seemingly simple string of numbers so difficult to nail consistently?
Vince Picone retweeted
Today, @NianticSpatial released an update to their AR companion, Dot, at Snap's Lens Fest, with new voice capabilities powered by Hume AI.
Dot's new interactive dialogue capabilities allow the AI companion to guide users through physical spaces, offering contextual information through natural, emotionally responsive dialogue.
Our CEO @AlanCowen, sees the product as a glimpse into the future of computing. “At some point, everyone is going to have AR in their lives in some format,” Cowen says. “You’ll be talking to companions that help guide you through the world.” Read more in today's @WIRED and article below!
Vince Picone retweeted
Introducing Octave 2: our next-generation multilingual text-to-speech model
What’s new:
- Fluent in 11+ languages
- 40% faster (<200ms latency) & 50% cheaper than Octave 1
- Multi-speaker conversation
- More reliable pronunciation
- New voice conversion & phoneme editing capabilities
For the month of October, we’re offering 50% off our Creator plan - use code OCTAVE2 at checkout!
Vince Picone retweeted
SambaNova 🤝 @Hume_AI
Imagine if the world's most realistic voice AI was also super fast, ridiculously smart, and crazy affordable 🤯
Well, we made it happen... and you can see it (& hear it 😉) for yourself👇
Vince Picone retweeted
What does empathy sound like in AI? 🤔
@Hume_AI’s hyperrealistic speech models now run on SambaNova, delivering voice AI that’s fast enough for real conversation (100–300 ms). 🏎️💨
Read the case study 👇
Vince Picone retweeted
2024: Voice Cloning
2025: What about personality cloning?
Hume’s voice AI can now not only mimic your voice but also speaking style and language.
It’s now available via our TTS and new speech-to-speech model, EVI 3, which is also launching today.
Vince Picone retweeted
We’re looking for a product designer to join us @hume_ai. You’ll help shape how the world interacts with expressive, real-time, emotionally intelligent AI.
Vince Picone retweeted
Meet EVI 3, another step toward general voice intelligence.
EVI 3 is a speech-language model that can understand and generate any human voice, not just a handful of speakers. With this broader voice intelligence comes greater expressiveness and a deeper understanding of tune, rhythm, timbre, and speaking style.
Vince Picone retweeted
There’s no “good” config file. All config is compromise. All user input is error.
That said, it’s JSON with comments for now.
Vince Picone retweeted
Today, we’re releasing Octave: the first LLM built for text-to-speech.
🎨Design any voice with a prompt
🎬 Give acting instructions to control emotion and delivery (sarcasm, whispering, etc.)
🛠️Produce long-form content on our Creator Studio
Unlike traditional TTS that just “reads” words aloud, Octave understands how meaning affects delivery to generate emotional, human-like speech.











