Mustafa Işık

Research at @synthesiaIO. Previously at @tavus @hedra_labs, @eth_en, @AdobeResearch and @TU_Muenchen.

[Wojciech Zielonka @w_zielonka]

After some extra time and additional work, I am happy to announce that D3GA will be presented at 3DV2025 in Singapore. The code and dataset can be found below. 🚀🚀

https://t.co/4Abz0CB0Xg
https://t.co/Baw1AF4jPK

[Hedra @hedra_labs]

In case you missed it... we dropped 5 HUGE updates this week 🔥

1. Voice Isolation – Remove background noise and get clear audio in seconds.
2. Voice Changer – Upload a voice or use cloned/custom voices and transform them instantly.
3. Voice Design – Generate unique voices from a simple text prompt.
4. AI Creative Assistant – Generate, translate, and enhance scripts faster.
5. 12-Minute Generations - Generate up to 12 minutes of content.

All these features are live now—don’t wait, try them today!

[Hedra @hedra_labs]

Character-2 is here!

A new generational leap for audio to video foundation models, and our next step towards a full-stack character model capable of intelligence, emotion, and more.

Key highlights:

* Widescreen and vertical support: removing limitations on aspect ratio.

* A massive quality boost: improved clarity, more expressive, without sacrificing our class-leading speed. Generate 4 minute long videos today.

Free today, link below!

[Mustafa Işık @realMustafaIsik]

@jon_barron Also, even way before 3DGS, Lombardi et al. actually did ray trace Gaussians wrapped around bboxes in the context of avatars but there was no notion of 3DGS around that time. It’s a brilliant paper: https://t.co/eKpJElyCc8

[Mustafa Işık @realMustafaIsik]

@janusch_patas Christmas came early for you!! You seem really happy 😄 looking forward to what you’re going to do with it

[Mustafa Işık @realMustafaIsik]

@Gabepluguez This is utter bs. The most important thing is caloric deficit for weight loss. There are numerous benefits of both resistance training and zone 2 cardio for your healthspan. Do not compare things in such a narrow context and mislead people!

[Hongwei Yi @HongweiYi2]

We are hiring Research Scientists/Engineers in Generative AI, especially with a focus on image and video generation! Come and join us to develop a cutting-edge video generation model with flexible multi-modal control.
https://t.co/Y8IL4M4OtG

[Hedra @hedra_labs]

Today we are releasing Character-1.5 and Stylize. Available now at https://t.co/mWS1TAkrQK.

This update introduces stylization, allowing you to transform into a character while maintaining your identity. Character-1.5 brings sharper visual output, more lifelike animations with natural blinking and head movements, and improved handling of uploaded images. It's a major leap forward for our foundation model before we release Character-2.

[Hedra @hedra_labs]

We’re thrilled to announce our $10M raise by @IndexVentures @speedrun @AbstractVC to build the next generation of storytelling technology to empower content creators.

At Hedra, we’re making the future of creative video with a multimodal studio that complements our foundation model, character-1. Our goal is not just to streamline content creation — but to empower brands and creators to captivate audiences in new, accessible ways.

We’re grateful to our investors and excited to invite creators from around the world to join us. We can’t wait to see what you make!

https://t.co/ud214y4sLW

[#CVPR2026 @CVPR]

The #CVPR2024 tutorial videos are now available!

https://t.co/COrwEHbR89

[Hedra @hedra_labs]

Thank you so much for the coverage @CharlieFink @Forbes! We can't wait to show you what's next.

[Hedra @hedra_labs]

Developers, developers, developers. Our API sign up is here! Please fill out this google form.

https://t.co/YF1Zdp9k2d

We're going to be onboarding you starting tonight, we will be prioritizing based on budget/hours of video.

Thanks for your patience as we work to scale Hedra to enable video for everyone.

[Mustafa Işık @realMustafaIsik]

🤯

[Hungry Donkey 🥕 @hungrydonkey]

🔥 My First AI Music Video!

"Pink Boy"

Image: @midjourney
Video: @LumaLabsAI
Lip-Sync: @hedra_labs
Music: @suno_ai_

Check out this AI-created video I made, and in the next tweets, I'll share my learnings and conclusions.

🧵 1/5

[Michael Lingelbach @mjlbach]

We launched ~48 hours ago, and have 10s of thousands of users producing over 100k videos. We can't wait to see what you create next as we build more of our platform, and we're just getting started :)

[Uncanny Harry AI @Uncanny_Harry]

Meet my mate Dave, he has a belated father’s day message for everyone, sound on please. I made Dave on father’s day with just a midjourney image, an eleven labs voice2voice voice and a new tool called HEDRA @hedra_labs .This is the closest thing to acting that I’ve seen from an AI generated video

[Mustafa Işık @realMustafaIsik]

You can visit https://t.co/8GfwyHTzVX to see what I used to generate the video below.

[Mustafa Işık @realMustafaIsik]

Finally out of stealth! Check the research preview of what our team has been working on recently. It's free now and a lot fun to use!

https://t.co/X7sQcE8D61

[Hedra @hedra_labs]

Introducing the research preview of our foundation model, Character-1. Available today at https://t.co/7FSVJiOlWS (on desktop and mobile).

* Infinite duration (30s for open preview)
* 90s generated per 60s (if our H100 supply holds)
* Expressive talking, singing, rapping characters

This is the first step in Hedra’s mission to build a multimodal creation studio accessible for everyone, giving creators complete control over emotional dialogue, movement, and (yes) entire worlds.

Pro tips in thread:

[Mustafa Işık @realMustafaIsik]

@janusch_patas Thank god, our jobs are safe :P

[Mustafa Işık @realMustafaIsik]

Holy shit 🤯

[OpenAI @OpenAI]

Introducing Sora, our text-to-video model.

Sora can create videos of up to 60 seconds featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions.

https://t.co/YYpOAcrXQ3

Prompt: “Beautiful, snowy Tokyo city is bustling. The camera moves through the bustling city street, following several people enjoying the beautiful snowy weather and shopping at nearby stalls. Gorgeous sakura petals are flying through the wind along with snowflakes.”

[Mustafa Işık @realMustafaIsik]

The most impressive results I’ve seen recently on animatable humans. It’s also really nice to see that ActorsHQ dataset is used here 🎊

[Yebin Liu @YebinLiu]

Animatable Gaussians: Learning Pose-dependent Gaussian Maps for High-fidelity Human Avatar Modeling
Projectpage: https://t.co/EChcKImrVb