We’re a team of artists, engineers, and researchers building controllable AI video editing tools to unbound human creative potential. Our research team builds AI video models to understand and affect fine-grained, controllable edits on any human in any video. Our product team makes these models accessible to everyone from everyday people, professional editors and animators, developers, and businesses large or small to reach new audiences and tell outstanding stories. Our technology is used to automate lip-dubbing in localization processes in entertainment, create dynamic marketing campaigns personalized to individuals or communities, animate new characters to life in minutes instead of days, fix mistakes in post-production by making word-level edits to avoid rerecording entire scenes, and more – this is just the beginning. In the last 12 months we graduated at the top of our YC batch (W24), raised a $5.5M seed backed by GV, won the AI grant from Nat Friedman and Daniel Gross, scaled to millions in revenue – and we're only just getting started. For more information, visit https://synclabs.so/
sync.的外部链接
US，California，San Francisco，94111
sync.转发了
It’s crazy to think how far we've come with video AI in such a short time. Check out lipsync-2 in action -- our latest model that preserves a speaker's unique style without any training. Huge shoutout to the sync. research team for their outstanding work. ???? It's amusing that soon even this breakthrough will seem rudimentary. Exciting times ahead! DM me for early access.
??
?? ???????????????????? ??????:: 36 Hours, 500+ Competitors… and We Took 2nd Place! ?? What a weekend. 109 teams. Non-stop coding. Intense competition. And out of all of it, we built something game-changing—MockMade, an AI-powered interview prep platform. ?????? ?????????????? ???? ?????? Interview prep is broken. It lacks real-world simulation, structured feedback, and adaptive coaching. So we built MockMade to change that. ?? AI-Driven Behavioral Interviews – Get real-time calls from an AI interviewer, mimicking actual interview conditions. ?? Smart Feedback & Scoring – Receive insights on communication, problem-solving, and structure. ?? Interactive Coding Interviews – AI-powered hints and adaptive feedback to sharpen problem-solving skills. ?????? ???? ?????????? ????: We combined cutting-edge AI and cloud technologies to create an immersive experience: ? Retell AI – Real-time AI-driven phone calls ? OpenAI GPT-4o – Response analysis & scoring ? Cartesia Sonic – Lifelike AI-generated voices ? Whisper – Audio transcription for accurate evaluation ? ide.js – Interactive coding challenges ? sync. – AI-generated lip-sync video simulation ? Full-stack architecture – MongoDB, Next.js, React.js, TypeScript ???????????? 2???? ??????????, ???? ???????? ??????: ?? Best Cloud Implementation – Sponsored by Caterpillar Inc. ?? Best Use of Sonic – Sponsored by Cartesia ($1000 in API credits!) ?? ?????????????? ???????????????? Huge thanks to my incredible teammates: Daniel Xie, Monish Muralicharan, and Aditya Agarwal – your creativity and technical brilliance made this possible. ????????’?? ???????? ?????? ????????????????? We’re not stopping here. Our roadmap includes: ? Smarter AI interviewer with adaptive follow-ups ? Expanding the interview database to cover more industries ? Launching MockMade to help job seekers worldwide ?? Check out our demo: https://lnkd.in/gf36QzjV ?? Join the waitlist: https://mock.meriedith.com This is just the beginning. Stay tuned! ?? #BoilerMakeXII #Hackathon #MockMade #AI #TechInnovation #InterviewPrep #PurdueUniversity
??
??->?? The past two weekends have been a nonstop grind, competing at TreeHacks, the biggest hackathon in the US, and BoilerMake XII the biggest hackathon in the Midwest ???? At TreeHacks??, my team and I created Anim-Education, an app that allows you to experience more fun and interactive lectures, using Meta Quest VR, OpenAI, and Groq. ?? Vision:? You join a Zoom meeting in VR, and our system transcribes the audio in real time. Then, using Groq’s?AI inference and Manim, we dynamically generate animations based on key concepts—turning boring lectures into visually engaging experiences. ?? Tech Stack: ? Groq API (Natural Language Processing + Animation Triggering) ? OpenAI (AI-powered analysis & generation) ? Manim (Auto-generated animations) ? Meta Quest (VR integration) Check out our project: https://lnkd.in/gTxwTYDD Meeting so many cracked students and incredible founders, hearing talks from leading figures like John Hennessy, driving an autonomous Cybertruck, and petting llamas ??, TreeHacks was one of the most exciting learning experiences I've had. Huge thank you to the TreeHacks organizers, sponsors, and my amazing team: Kathie Huang, Matthew Steele, and Tyler Kwok for an unforgettable experience and for showing me around SF ??! At BoilerMake Manav Gagvani, Kathleen Young, Robert Zhang, and I won sync.'s best use of lipsync, with slynk: an immersive ad experience using AR. With slynk, scanning an ad transforms it into an immersive, personalized experience.? ?? What it does: Scan an ad → A personalized 3D spokesperson comes to life, walking you through the product, answering your questions, and showing how it fits into your life—all in real-time AR. ?? Tech Stack: ? ARKit ? LLaMA Vision (AI-powered video inference) ? ElevenLabs (Text-to-speech) ? sync. (Lip-synced video generation) Watch our demo if you wanna see LeBron selling you a Laker’s ticket: https://lnkd.in/gtD3U7xD Devpost: https://lnkd.in/gPn7td_Y Thanks to my team and the BoilerMake team for putting on a great event! Can't wait to meet the sync. team in SF ?? #treehacks #boilermake
great job ! keep going ??
We won second place at Purdue University‘s Boilermake—the largest hackathon in the Midwest! In 36 hours of hacking, we built a product that won 3 prizes, including 2nd best cloud implementation by Caterpillar Inc. and best use of Sonic by Cartesia, securing over 2,500$ in prizes among 109 teams and 500+ participants! I couldn’t have done it without my talented and hardworking team, Daniel Xie, Irfan Firosh, and Aditya Agarwal. We built MockMade, an all-in-one interview prep platform using human-like real-time voice to simulate behavioral and technical interviews. Our behavioral interview is facilitated by a call to the candidate’s phone equipped with an AI interviewer agent and an in-depth transcript analysis post-interview. Our technical interview is a leetcode-like IDE integrated with an AI agent to guide you through the problem to get to the final answer. We used Retell AI to power the physical phone call with a human-like voice agent and OpenAI GPT-4o to analyze the interview in various different categories. We generated an AI lip sync video simulation of the phone interview using sync. For the technical interview, we used ide.js to create the leetcode environment, Whisper to transcribe the audio, GPT to generate a response, and Cartesia to simulate human-like speaking. Special thanks to BoilerMake, Caterpillar, and Cartesia for the recognition and prizes. I personally enjoyed pitching at the Shark Tank to the professor judges alongside the top 8 teams. All the cool people I met and new technologies I worked with made this weekend an unforgettable experience! We're excited to see what's next for MockMade! Check out our product and demo: https://lnkd.in/gf36QzjV Sign up for our waitlist here: https://mock.meriedith.com #BoilermakeXII #Hackathon #MockMade #Innovation #PurdueUniversity #AI #Interview
glad to be of help ! keep going ??
CS@Purdue | Computer Vision Researcher @ OpenInterX | Incoming External ML Research Intern @Peraton | Machine learning engineer and quantum computing enthusiast — passionate about launching innovative ventures.
Welcome to Verbatim — Understand Your Videos, Verbatim. Last weekend at BoilerMake, our team built Verbatim, a video understanding platform that goes beyond simple transcription. We didn’t win, but our frustration turned into determination—and we took our hackathon project further. What Verbatim Does: ?? Translates speech in videos into 20+ languages ??Summarizes content for quick insights ???Creates a digital clone that lip-syncs perfectly in the target language in the original speaker's voice ??Uses memory-level video embeddings for real-time Q&A and topic search This wouldn’t have been possible without my incredible team: ?? Pranav Neti – Seamlessly integrated frontend & backend (plus an awesome loading gif) ?? Sonny Chen – Powered through deployment struggles to launch on Vercel ?? Cindy Yang – Built the database & authentication, and coined the name Verbatim! I'd also like to mention the innovative startups that helped power our project: the team at sync., who supported us through the night, and TwelveLabs, whose video memory allowed for speedy Q&A. ?See Verbatim in action: https://lnkd.in/gH55_HJJ ??? Try it out: https://lnkd.in/g5pakASP
Here's what's new in the month of feb: – We shipped a new way to organize your videos, introducing projects to the sync.platform – You can now estimate how much an api call will cost you with our new /cost-endpoint – We've 10x'd our compute capacity for more throughput, generation speeds are now up ~15% across all our models and we can serve a higher capacity We're a company building tools to help people put themselves on camera, so our team put themselves on camera :) Check out our latest changelog to see what else is new ??
sync.转发了
welcome back to office ?? the?sync.?team is back in the hq, which means... ?? constant, never ending table tennis games ?? more $100/day food deliveries ??? standups every hour but it also means we're shipping, every single day. this week, the team shipped teams - a way to finally collaborate on sync. and I spent the week recording the team shipping teams, lol. check it out → https://lnkd.in/gs5vf8hh
welcome to the team Sebastian Semeniuc ??
I’m happy to share that I’m starting a new position as Product Fullstack Engineer at sync.!
此处无法显示此内容
在领英 APP 中访问此内容等
sync.转发了
We’re excited to announce our partnership with sync. to bring their newly released 1.9.0-beta model for zero-shot lipsync natively into the Sieve ecosystem. This model marks a step change in zero-shot realism and is now available as a backend option in Sieve’s lipsync and dubbing pipelines. This is the first of much more to come working with Prady Modukuru & team. We can’t wait to see how developers use this combination to build creative and social experiences that break down barriers in content creation and language. Read more about the partnership via the blog in comments.