Loading video...

Video Failed to Load

Go Home

The next 5 years will be wild. Knowing how to code and how these AI models work will give you infinite leverage. We integrated OpenAI's Whisper with Spot. We can now control the robot using our voice! Nothing is more rewarding than seeing your code walk on four legs.

277,558 views • 3 years ago •via X (Twitter)

38 Comments

Santiago's profile picture
Santiago3 years ago

We capture audio and transcribe it using Whisper. It's fast. We can react quickly to commands. We then map the transcription to individual functions that Spot understands using its SDK. It feels magic having the robot reacting to your voice!

Santiago's profile picture
Santiago3 years ago

This is just the start. The team @Levatas has been working with robots for a long time. Plans for the next year alone are crazy! Stay tuned. I'm working on a couple more experiments. They are wild. How long before robots become a big part of our lives?

Jack Forge's profile picture
Jack Forge3 years ago

@OpenAI I have 2 Spots at work. I’m scared of them enough as it is. 😆

Santiago's profile picture
Santiago3 years ago

@OpenAI Lol

Prince Canuma's profile picture
Prince Canuma3 years ago

@OpenAI This is a very cool application! 🚀 I can imagine trigger words to multi-step actions. Go get me a beer 🍻😜 Undock > locate fridge > open it > locate beer…

Chris Nielsen's profile picture
Chris Nielsen3 years ago

@OpenAI Using the world’s most advanced technologies to unlock human potential Love it!

Raul Junco's profile picture
Raul Junco3 years ago

@OpenAI Nice integration, @svpino! Say Hi to David and the rest of the team!

Deisbel (Sr. Software Developer)'s profile picture
Deisbel (Sr. Software Developer)3 years ago

@OpenAI Happy to see the progress you are making integrating Spot with new communication ways. I still remember my son playing with Spot “the pet” in our offices at Levatas 5-6 years ago. Congrats.

Gus (🤖🧠+🐍+🥑🗣️)'s profile picture
Gus (🤖🧠+🐍+🥑🗣️)3 years ago

@OpenAI That's very cool! Can the bot do complex commands like: "Undock then go measure something and then dock back"? or it does only short ones and you have to control the sequence?

Santiago's profile picture
Santiago3 years ago

@OpenAI Right now, only short commands. Not too hard to add support for multi-step commands.

Passenger's profile picture
Passenger3 years ago

@OpenAI I would say that GPT 4 knows how to code this. So, it is a question of to want and have the means to do what you want.

Multimodal's profile picture
Multimodal3 years ago

@OpenAI Eventually we can even control it with our minds! Awesome work!

Aditya Thakur's profile picture
Aditya Thakur3 years ago

@OpenAI Wow this is so cool 🚀

Dayo Jaiye 👑's profile picture
Dayo Jaiye 👑3 years ago

@OpenAI Nice 👍

essa's profile picture
essa3 years ago

@OpenAI @Susejlava

Hector J.C.'s profile picture
Hector J.C.3 years ago

@OpenAI Me imagino un robot similar ayudando a astronautas en la luna al que simplemente hay que darle instrucciones de voz

Sergio Manrique's profile picture
Sergio Manrique3 years ago

@OpenAI Very nice integration ! I wonder how this could be implemented at scale !

ML Neurodivergent's profile picture
ML Neurodivergent3 years ago

@OpenAI Could sound stupid but could it misunderstand a command and obey right away something like “hit that human”?

Arinze's profile picture
Arinze3 years ago

@OpenAI Wow

Dr. Brian Massey's profile picture
Dr. Brian Massey3 years ago

@OpenAI Love this! The downstream applications of this are as exciting as they are numerous!

Daniel's profile picture
Daniel3 years ago

@OpenAI Pretty coool

Marc M.'s profile picture
Marc M.3 years ago

@OpenAI Really cool implementation of Whisper. Nice to see it applied to hardware and not just a terminal/IDE output. 🤖

Gabriel Osorio's profile picture
Gabriel Osorio3 years ago

Now they listen, and according to this other work, with LLMs they can see and understand better. We are witnessing a revolution unlike any other in history: fast and global. The future is fascinating, may good fortune always accompany humanity to walk wise paths.

Lawrance K's profile picture
Lawrance K3 years ago

@OpenAI Progress!

Aishwarya Nevrekar's profile picture
Aishwarya Nevrekar3 years ago

@OpenAI Awesome 😎

Terry Palmieri's profile picture
Terry Palmieri3 years ago

@OpenAI Nice

Anthony Richard's profile picture
Anthony Richard3 years ago

@OpenAI Incredible. Remaining visual or motion commands for victims who need rescue and can not utter a word, some sort of life signature for spot to read and act on. Congratulations for this level.

TnvMadhav's profile picture
TnvMadhav3 years ago

@OpenAI We can now communicate with drones using a walkie talkie?

Pritesh Singh's profile picture
Pritesh Singh3 years ago

@OpenAI Wow!

Princeofpalmair's profile picture
Princeofpalmair3 years ago

@OpenAI I'm a retired engineer of 40 years. I'm a lifelong s. Floridian, I want to learn actual AI, what is the best educational tool you can recommend, I appreciate it. I have some fantastic virtual applications, ( in my cloud)

G's profile picture
G3 years ago

@OpenAI @sendvidbot

Bananstaben's profile picture
Bananstaben3 years ago

@OpenAI i can think about a few things more rewarding

VuDuong's profile picture
VuDuong3 years ago

@OpenAI imagine that dog have a gun.

Stefaan Tysebaert's profile picture
Stefaan Tysebaert3 years ago

@OpenAI So who had the emergence of skynet for 2023?

Avik Dutta's profile picture
Avik Dutta3 years ago

@OpenAI @BiswajoyChatte1 You did something like this several years ago with a laptop in the front...

thom's profile picture
thom3 years ago

@OpenAI @readwise save thread

kox's profile picture
kox3 years ago

@OpenAI Have you used GPT-3.5 or GPT-4 to translate natural language into spot sdk commands in this video?

Places Visited & Pictures Taken's profile picture
Places Visited & Pictures Taken2 years ago

What if we use AI? Let’s find out.

Related Videos

I'm teaching a new course! AI Python for Beginners is a series of four short courses that teach anyone to code, regardless of current technical skill. We are offering these courses free for a limited time. Generative AI is transforming coding. This course teaches coding in a way that’s aligned with where the field is going, rather than where it has been: (1) AI as a Coding Companion. Experienced coders are using AI to help write snippets of code, debug code, and the like. We embrace this approach and describe best-practices for coding with a chatbot. Throughout the course, you'll have access to an AI chatbot that will be your own coding companion that can assist you every step of the way as you code. (2) Learning by Building AI Applications. You'll write code that interacts with large language models to quickly create fun applications to customize poems, write recipes, and manage a to-do list. This hands-on approach helps you see how writing code that calls on powerful AI models will make you more effective in your work and personal projects. With this approach, beginning programmers can learn to do useful things with code far faster than they could have even a year ago. Knowing a little bit of coding is increasingly helping people in job roles other than software engineers. For example, I've seen a marketing professional write code to download web pages and use generative AI to derive insights; a reporter write code to flag important stories; and an investor automate the initial drafts of contracts. With this course you’ll be equipped to automate repetitive tasks, analyze data more efficiently, and leverage AI to enhance your productivity. If you are already an experienced developer, please help me spread the word and encourage your non-developer friends to learn a little bit of coding. I hope you'll check out the first two short courses here!

Andrew Ng

1,224,713 views • 2 years ago