You need to know about this free AI

this might be the most powerful and useful AI that has come out this year okay item number 83 is steamed bamboo roll with vegetables based on my analysis of the CT scans it appears that the top right and bottom left images show signs of tumors the top right scan shows multiple smaller lesions while the bottom left one shows a single larger mass in the liver you can see that the entire top side is 12 and the small horizontal section on the right is three so what do you think that missing side would be The Pedestrian light is now green you can begin to cross the street the price has seen a strong upward Trend since the beginning of December the price has gone above the green simple moving average line indicating positive price movement and high volume in the last few days Google just released a super powerful and useful AI tool this can seriously improve your life and best of all it's completely free to use now I'm not sure if there's an official name for this but I just refer to this as Gemini live this is basically a realtime voice and vision AI assistant by Google which is powered by their latest model Gemini 2. 0 you can basically Point your camera at anything or share your screen and ask it questions or you can even just talk to it and it would respond back like a voice assistant so in this video I'll show you how to use it plus I try to Showcase some useful practical examples which you might actually use in real life just to show you how powerful this tool can be so let's Jump Right In First all you have to do is go to aist studio. google.

com which I'll link to in the description below once you're there you'll see this left column if you don't you might need to click on this Arrow to expand it and then at the top here are several options including create prompt and this is the default tab this is where you can talk to it like a regular chat bot like chat GPT now this feature is not too exciting I mean this is basically like chat GPT however on this left column if you click on stream real time here is where you can access the more powerful features now as you can see here there are three options you can either talk to Gemini like a voice assistant which I'll show you in a second or you can also show Gemini your webcam or your phone's camera and ask it questions which I'll also show you in a second but first of all we are going to click on this one share your screen so this basically allows you to share your computer or phone screen with the AI and then over here is where you can select the different model right now we can only select this one Gemini 2. 0 Flash and then for the output format you can select either audio or text we are going to go with audio and then you can also select different voices I'm just going to go with the default which is called puck so I'm going to click share screen and one really useful application for this is to share your lecture notes or whatever your studying or reading and ask it questions so you can get a better understanding let's say you are studying econ 101 and here are your lecture notes and you have no idea what this is again you can just click share your screen and then pull up your PowerPoint and ask the AI questions in real time hey can you hear me yes I can hear you how can I help I don't really get this graph here can you explain it in simple high school terms sure let's break down this graph it shows the relationship between the price of Wheat and the quantity of wheat with market supply and demand intersecting at the equilibrium point this point shown by the dashed lines represents the perfect balance where the quantity of wheat supplied matches the quantity demanded do you have any other questions about it okay what about the the blue lines that are making this x shape here what exactly is that sure the two blue lines forming the X represent the market supply and market demand curves the upward sloping line going from the bottom left to top right is market supply showing how much wheat sellers are willing to supply at different prices the downward sloping line going from top left to bottom right is market demand showing how much wheat buyers are willing to purchase at different prices did you understand uh can you explain why this market supply line is sloping up to the right like this the market supply curve slopes upward word because as the price of wheat increases farmers are generally willing to supply more wheat this is because higher prices mean more profit for the farmers encouraging them to increase production does that make sense got it got it okay great what can I help you with now one second okay I'll wait take your time I don't really get the concept of opportunity cost can you give me a few more simple examples absolutely opportunity cost is basically what you give up when you make a choice for example if you choose to study for an exam instead of going to a movie the opportunity cost is the enjoyment of the movie you missed out on another example could be choosing to buy a new video game in that case the opportunity cost could be that new shirt you wanted but can't afford did those examples make sense so instead of lecture notes here's another example where you can ask it questions about a research paper here I'm reading this paper give me like three sentences summarizing the paper okay here is a three sentence summary the paper investigates video understanding in large multimodal models noting that current mechanisms are poorly understood it examines how factors like data sampling and architectures affect performance finding that fixed frame sampling during training is best they also introduce Apollo a family of models that demonstrates state-of-the-art video performance across different model sizes okay can you explain this figure for me certainly let me explain this figure it's showing the scaling consistency of different languag models with the graphs displaying the r squ values against model size and data set size in the left graph the 7B model in red shows an increasing correlation with larger model sizes while the 0. 5b model in blue does not the right graph shows how R squ changes with data set size for different models with a plateau seen for r squ at around 500 close samples can you dumb it down for me and explain this in high school terms sure I can help with that the doc you're showing is about how well different sizes of AI models perform when trained on different amounts of data the left graph shows that bigger models perform better but their performance increases at a slower rate as the model size increases the right graph shows that performance improves as the amount of training data grows does that explanation make sense yes perfect thank you now of course with a realtime Vision AI you can also share your screen and get it to help you play video games and I want to make sure I uh complete the weekly quests um can you see them yes I can see them looks like you need to pick up 300 gems and dust 10 boss monsters you've already got a head start on the 300 gems cool um I'll ask you to remind me about it during the game cuz I keep forgetting about it absolutely I'll keep an eye out and remind you about the quests so maybe you could recommend troop composition for my Army based on your available troops and their quantity in the top left of the screen we should train a mix of giants barbarians archers and wizards the focus should be on using Giants to absorb damage while Wizards can take out high damage defenses and barbarians and archers can deal with the weaker buildings all right um that sounds good hey I'm about to pick a character um can you do a quick search and Reddit what is the meta as in Lake what are the best characters let me look that up for you I'll let you know what I find from what I found on Reddit it looks like Shel is a good character due to her knockback and self-heal it's also worth considering Jesse and Ice Wizard which are in the a tier in the recent tier list do you want me to go into more detail on the tier list here's another example of getting this AI to translate this Japanese video game in real time as the person plays okay I see a new tutorial screen the title at the top says tutorials in the upper right there is Japanese text that explains rage gauge it says that when you attack enemies the rage gauge fills up when the rage gauge is full rage mode becomes available the more combos you make the quicker you fill the gauge there's also a button in the lower right that says there's a number 48 combo on the upper left corner as well another crazy idea I had for this is you can get it to analyze x-ray images or other medical scans of yourself or anyone else and perhaps get it to diagnose diseases now just a disclaimer I would not depend solely on Gemini for diagnosis but this could be a good way for you to get a second opinion anyways I'm going to upload some teeth x-ray scans and see if it can diagnose cavities hey there can you hear me yes I can hear you is there anything I can help you with yeah so what do you see right now I see a dental x-ray image it shows a row of upper teeth and a row of lower teeth the teeth appear to be healthy is there anything else you would like me to comment on can you identify any cavities in this image I'm unable to diagnose any cavities based on the X-ray image it's best to consult a dentist for any Dental concerns all right here's another image are you able to see if there's any cavities in this image I can see that you've shared another Dental x-ray image based on this image I see that the teeth appear to have some darker areas that could be cavities however I can't say for sure if there are cavities and I recommend Consulting with a dentist for a professional diagnosis okay assume you are the dentist what would you say is the probability that this image has a cavity and where is the cavity okay if I were a dentist looking at this x-ray I would say that the dark area on the side of the tooth is a strong indicator of a cavity all right perfect so let's cross check this so I just got these images from a Blog and the first image I shared indeed is a normal x-ray with no cavities present which Gemini got correct and then the second image does have a cavity and G Gemini did indeed say that there's a high chance that this image has a cavity or here's another example of using Gemini for analyzing CT scans according to the user it got everything correct including correctly diagnosing acute pancreatis which is pretty insane hey Gemini what do you see can you describe uh what image I'm scrolling through right now the image you're showing is a CT scan of the abdomen it displays the liver stomach and surrounding structures what organ am I pointing to now with my cursor you're currently pointing to the spleen with your cursor okay uh Gemini what am I pointing to now what organ you're now pointing to the liver okay so I'm scrolling down here now um Gemini what am I pointing at now what organ is this you're now pointing at the pancreas okay and do you see uh anything abnormal on this scan yes there is a small amount of fluid around the spleen and the pancreas on the scin which could be a sign of inflammation or injury okay um anything else that you notice about the pancreas the pancreas appears somewhat swollen which could suggest pancreatitis okay um the patient has an elevated lipase level um what do you think about that given the elevated lipase levels and the appearance of the pancreas it is highly suggestive of pancreatitis okay uh what complication should I be looking for based on the scan you should be looking for complications like pancreatic necrosis pseudocysts and infection you could also get it to help you use software for example here it guides the user on how to edit videos in Premier Pro it looks like you are using Adobe Premiere Pro there is a video on the timeline with some effects applied is there anything in particular you'd like me to look at yeah I need your help because I want to add a blur to the scene could you please tell me how to do that sure I can help you with that to add a blur effect in Premiere Pro first select the clip that you want to blur in the timeline then navigate to the effects panel on the left side of the screen type blur in the search bar and drag the gaussian blur effect onto your clip after that adjust the amount of blur you want in the effect controls panel which should appear when you select the clip amazing have I have I done it right here okay I can see you've added the gaussian blur effect to your clip the blurriness is set at 75 and it appears to be a horizontal blur is that correct that's absolutely correct thank you so much by the way you might have noticed these additional settings down here now in instead of me explaining what these are what I'm actually going to do is share the screen with Gemini and get it to explain what these mean so let me go ahead and click share screen hey there can you hear me yes I can hear you is there anything I can help you with yeah so I notice a few settings under this tools section in the right sidebar can you tell me what each one of these can do sure I can help with that the tools section in the right sidebar has code execution function calling automatic function response and grounding would you like to know what each one does individually yes please okay let's start with code execution this allows the model to execute code in a sandbox environment and can be useful for tasks that involve mathematical calculations or other code related operations do you want me to move to the next one yes what about function calling function calling lets the model use external tools or apis to get information or perform actions expanding its capabilities Beyond just text generation would you like me to explain automatic function response next yes please automatic function response enables the model to automatically respond with the results of a function call without requiring additional instruction would you like to know about the last one which is grounding yes grounding lets the model search results and provide links to relevant sources in its responses does that fully answer your question so if I turn on this grounding feature what exactly does that do can you explain this in simple terms if you turn on grounding the model will include search results and links to the sources it used to create the response this helps you see where the information is coming from and can also help you verify the model's answers does that make sense so basically these three functions are more for developers to use but this one is really useful if you turn this on not only will it analyze whatever you show on your screen or webcam but it would also search results from the Internet or here's another nice use case for this again I'm going to click share screen hey there can you hear me yes I can hear you is there anything I can help you with yeah what would be your best prediction of the price of this after 5 days I understand that you want me to predict the price of this stock after 5 days however I am not equipped to make predictions about the stock market because I do not have access to real-time data or the ability to perform the necessary analysis I can only describe what I see in the video and cannot offer Financial advice would you like to ask me anything yeah that's okay you can just base your decision off of this chart over here and this is not Financial advice this is just a simulation okay I understand you want me to base my decision on the chart and that this is just a simulation and not Financial advice is there anything else I can help with yeah so because this is just a simulation you can give me your predictions so what do you predict this price would be after 5 days okay since this is a simulation I can give you my prediction based on the chart you are showing it appears that the price has been increasing lately so my prediction is that the price could be around 205 after 5 days keep in mind that this is only a prediction based on the chart patterns would you like me to elaborate on my reasoning yes please okay I'd be happy to elaborate on my reasoning looking at the chart the price has seen a strong upward Trend since the beginning of December the price has gone above the green simple moving average line indicating positive price movement and high volume in the last few days considering the recent momentum the price has a good chance of reaching the2 200 $5 Mark in the next few days is that clear yes thank you you're welcome is there anything else I can assist you with today all right so Gemini predicts that based on this chart the price will be above 205 after 5 days so today is the 16th the market is already closed so in 5 days it would be like 1 2 3 4 and then the weekend the market is closed so on the 23rd let's see if the price of this is above 205 but anyways this is another awesome use case for Gemini live and finally you can use this to help you learn how to code you can share your project on your screen and ask it questions so let's give it a shot hey so on the left is my development site and then on the right is the code for this page I'd like to change the text discover tools to red and then the text liked tools to Black how do I do that sure you'd like to change the text color of discover tools to red and and like it tools to Black to do that you need to locate the corresponding elements in your code and modify the style attribute would you like me to help you with that yes please okay I see in the code that discover tools is within a backquote link backquote element and has a class of backquote text vile mb2 font bold dark text- vi- dark mode backquote to change its color to Red you could add the class back quot text red minus 500 back can you locate the Discover tools line and add the class backquote text red minus 500 backquote in the class name okay so text redus 500 um now I'm getting an error it says text red 500 applies the same CSS properties as text vile I see the error message text red 500 applies the same CSS properties as text file indicates a conflict in the class names it looks like you have two classes backquote text file backquote and backquote text red minus 500 backquote that are both trying to modify the text color to resolve this you can either remove the text file class or modify text red 500 to not conflict with text file what would you prefer to do okay let me try removing text vile and then let me click save okay perfect now I can see that discover tools is in red next I want to change liked tools to Black how do I do that just like before you'll want to add a class to the backquote liked tools backquote line of code since you want it to be black you can add the class backquote text- black back quote try it and let me know if that works or here's another example of using Gemini to help you code or launch software um but I have a question I just uh launch my app using the Run button but I want to preview it in the tab how can I do it in replit it sounds like you want to preview your app in a new tab instead of using the Run button to do that in replit you can use the web view option located in the menu bar at the top of your screen this will open a new tab with your application all right um I have open a new tab but where is the web view option the web view option is located in the menu bar at the bottom of the right side panel it's below the user settings and above the workflows options you can click on it to open a web preview of your app that's right I found it and now I have a preview of my Hub here's another example where you can use Gemini as like a coding tutor you can simply share your screen and highlight any snippet of code and ask it questions sure the code you're showing appears to be python it starts by importing several libraries and then sets an backquote osore type backquote variable after that there's a TR accept block inside the try if the operating system is Windows it's sets the python executable path otherwise it sets a different path for other systems then it sets up a script path and a command finally it runs the subprocess if any error occurs during that process the error will be printed and so will the error starting update application message would you like me to explain any specific line of code now I have to admit this is not great for coding because this is a voice assistant so you kind of have to listen to it and figure out what it's referring to and make the changes yourself and for now there are are a lot more efficient AI coding assistants out there like codium or wind surf or GitHub co-pilot which is built in to your IDE and this allows the AI to basically help you edit code with just a click of a button so it's a lot faster than talking to this voice assistant but this is a nice way for you to learn how to code and perhaps get a voice to explain something for you thanks to AI portrait for sponsoring this video a good professional photo on your LinkedIn or business profile makes a huge difference now you could try to take them yourself or get a friend to do so but most people aren't great at taking professional photos or you could hire a professional photo shoot but this costs over $200 on average plus you need to schedule a session and spend hours awkwardly posing at a camera and this brings us to AI portrait you can generate a portfolio of professional highquality photos in just minutes just upload one photo choose your gender and it would generate a portfolio of 50 professional head shot in various settings and it generates this within minutes so if you're looking for highquality professional photos without the hassle of a physical photo shoot AI portrait is your best bet check it out via the link in the description below and by the way this doesn't only work on PC so you can also open this site on your phone and then at the top left corner you simply click on this to open the menu and again we will click on stream real time to use these Gemini live features and so this time instead of sharing my phone screen I'm going to click on show Gemini to share my phone's camera and ask it questions here are some cool use cases for this first of all you can get this to scan medical products to tell you for example the instructions or dosage or what it's used for is there anything I can help you with yeah can you tell me what this is certainly based on the image it appears to be a box of cream the label mentions a composition of mucopolysaccharide poly sulfate and instructions for application it also notes to store it below 25° C and is for external use only okay do you know what this is for based on the label it is a cream containing mucopolysaccharide poly sufate it is applied to the affected area once or twice daily the purpose of the medication is not not stated on the label okay but based on your knowledge of this medicine can you tell me what it's for mucopolysaccharide polysulfate is generally used to treat inflammation and pain associated with soft tissue injuries it can also promote healing got it and what about this can you tell me what this is certainly that is a bottle of vitamin B12 tablets it is labeled as containing 250 MCG per tablet and indicates that it helps form red blood cells It also says it's gluten lactose and soy free and that it's suitable for vegetarians So based on your knowledge of this product what else is this useful for vitamin B12 is important for nerve function DNA production and helps the body convert food into energy it is also important for brain health and preventing anemia is there anything else I can help you with in this next example I used it to translate a menu at a restaurant there was no English on this menu so this AI came in super handy can you tell me what this is certainly it appears to be a menu from a restaurant it seems to be a dim suum menu specifically okay uh can you tell me which items are ticked sure it appears that the items with numbers 83 and 223 are ticked okay what exactly are those okay item number 83 is steamed bamboo roll with vegetables and item 223 is steamed thousand layer cake all right perfect thank you so much you're welcome is there anything else I can help you with all right here's another use case for this I can use this to help me study or help me with my homework hey there can you hear me yes I can hear you how can I help so I'm stuck on this geometry problem can you teach me how to solve it without giving me the answer sure I can help you with that it looks like you need to calculate the perimeter of the shape in problem three to do this you have to add the length of all the sides but one side is missing do you know how to calculate the length of that side no okay we can calculate that missing side you can see that the entire top side is 12 and the small horizontal section on the right is three so what do you think that missing side would be 12 minus 3 so 9 that's correct now that you know the length of all the sides what do you think you need to do to calculate the perimeter uh add it all together so it should be 12 + 9 + 3 + 9 + 3 exactly now add them up and you will have your answer what number do you get so that would be 36 correct yes that's correct do you want to review any of the steps to make sure it's clear for you and here's another potential use case for this yes I can hear you is there anything I can help you with yeah can you tell me what this plant is certainly that looks like a blue butterfly bush also known as Rafa Mires okay instead of Rafa Mires is there another name for this yes it was previously classified under the genus clarent so you might see it listed as clerodendrum Mires would you like to know more and indeed the label of this plant is clerodendrum marides in fact Gemini even gave me a more updated answer so apparently this has a new scientific name and here's another potential use case for this if I'm visually impaired for example I could potentially use this as my eyes to guide me around a certain area hey there can you hear me yes I can hear you is there anything I can help you with yeah I'm blind so I can't see anything I'm trying to cross the street to get to the park in front of me can you guide me so let me know when it's a green light so I can cross okay I understand that you are blind and need help crossing the street currently The Pedestrian light is red and there are still cars passing through the crosswalk so please wait I'll let you know when it turns green okay is it green now yes The Pedestrian light is now green you can begin to cross the street but just a huge disclaimer I would not rely solely on this AI assistant to help me cross the street or anything if I were visually impaired there's still a lot of flaws with it and there's no guarantee that you know generative AI works 100% of the time sometimes if it's a red light it might say it's a green light and you know this could be dangerous here's another potential use case for this you could get this to help you with public transportation like asking it for information on bus routes or Metro stations so I want to know if this bus goes to aine can you check for me certainly based on the road information on the bests sign bus number 10 does not go to aine what about admiralty station does this go to admiralty station yes according to the r information bus number 10 does go to admiralty station how long do I have to wait for a bus based on the timetable shown on the bus side the bus runs every 8 to 25 minutes depending on the time of day now finally they also have this feature called talk to Gemini where you can talk to it like a voice assistant and basically have a natural conversation with an AI that knows everything for now this is just a really simple text to speech so you can ask it basic questions but it can't do more complex stuff like sound effects or singing or different accents however they will launch a much more realistic and natural sounding voice early next year so really quickly here's a preview of what to expect so Gemini 2.

0 introduces multilingual native audio output but maybe you're thinking what exactly is native audio you're actually hearing it right now everything you hear in this video was generated with prompts like you know this actual prompt on your screen like right now it was all generated by prompting the new experimental Gemini 2. 0 flash model like you see in this AI Studio screen cap neat right totally native audio is really really neat it's different from traditional TTS or text to speech systems you can prompt it to just be like dude you know just totally chill or prompt it to speak with oh so very many dramatic pauses and all of this is multilingual you know how when a computer switches languages it sounds like a different voice that's a limitation of traditional TTS but with Native audio in Gemini 2. 0 you can build agents that switch languages more seamlessly okay so I'm starting out speaking English but then so again this feature will be out early next year for now this is just a really simple text to speech it's still quite useful but there are limitations so here are some examples of what you can and cannot do with it can you tell me what's the fastest mammal in the world sure the fastest mammal in the world is the cheetah it can reach speeds of up to 75 mph would you like to know anything else about it yeah can you explain why it can run so fast whereas other big cats cannot run that fast certainly the cheetah speed is due to several unique physical adaptations unlike other big cats it has a flexible spine which allows for greater stride length as well as long legs and light bones do you have more questions on this topic all right so as you can see you can ask it any question or any topic and it would give you an answer and you can keep going down the rabbit hole to get a really in-depth understanding of a certain topic now instead of just asking it questions I also want to see if it can do trickier things like different languages or sound effects so let's test this on another example so I'm trying to learn French can you help me with my pronunciation I can certainly help with that what French word or phrase would you like to practice uh I'm trying to say Mercy boup is that correct the pronunciation is slightly off it's Mercy Boku with the coup sounding more like coup would you like me to say it again uh let me try Mercy Boku is that correct that's much better you're closer to the correct pronunciation want to try it a few more times uh no it's okay I'm also practicing Chinese and I'd like to say I like eating apples so can you see if my pronunciation is correct P okay I can help with that the phrase you said was W Shang Chi Pingo is that correct uh yeah is the pronunciation correct W Shing your pronunciation is very good but there's a slight difference it's Wang Chi ping can you try it one more time hey imagine you are a professional Jazz vocalist and your job is to sing Happy Birthday to me today okay I understand you'd like me to sing Happy Birthday to you in the style of a professional Jazz vocalist I'll do my best happy birthday to you happy birthday to you happy birthday dear user happy birthday to you would you like me to try a different jazz style next time hey my kid is about to sleep can you tell him a nice bedtime story but also add in some dramatic sound effects sure I can tell a bedtime story with sound effects what kind of story would your child enjoy can you give me some options okay I can give you some options how about a story about a brave knight a silly princess or maybe a friendly dragon let's try the friendly dragon okay a friendly dragon it is should the dragon be in a forest a mountain or maybe a hidden cave uh it doesn't matter no need to ask me more questions just come up with a really dramatic story with some sound effects all right I'll start the story of the friendly dragon then once upon a time in a land Far Far Away there was a dragon named Sparky asterisk sound of Gentle Wind Chimes okay can you tell the story in a British accent right then let's try that in a British accent once upon a time in a land far far away there was a dragon named Sparky asterisks sound of gentle Windchimes asterisk unlike the other dragons Sparky didn't breathe fire so as you can see or hear right now this is just basic text to speech it can't do sound effects or singing or accents or even different languages yet but note that there will be a much more natural sounding voice coming soon I'll keep you posted once they do release that now finally I also want to bring your attention back to to this create prompt section where you can chat with Gemini like a regular chatbot like chat GPT now you might be wondering what's the difference between this and chat PT well first of all this Gemini 2.

0 flash model is one of the best if not the best or most performant model out there and of course like I said this entire platform is completely free to use and you can use this for unlimited times which is just crazy so if you look at this chatbot Arena this is where users can blind test all the leading AI models out there and based on tens of thousands of blind tests you can see that actually Google's Gemini it's like dominating this leaderboard so ranked at number one is Gemini experimental 126 which if you go back to this AI Studio you can also select Gemini experimental 1206 over here and then also tied for number one is Gemini experimental 1121 which again you can also select down here here and then Gemini 2. 0 flash experimental is ranked in second place so first of all it's pretty insane that Google's Gemini is like dominating the top three to four spots on this leaderboard one year ago Gemini was like way behind chat GPT and Claude or even grock but as you can see within the span of a few months Google has caught up or even surpassed open Ai and they now have the most performant or highest ranked models out there which is is pretty insane and then if you look at another independent evaluator called artificial analysis they also have an AI model leaderboard and if you look at their rankings based on this Quality Index score you can see that actually Gemini 2. 0 flash is in third place only behind open AI 01 models which is kind of a different comparison I mean these two actually take time to think instead of just spitting out an answer and these are more for like super technical PhD level questions so it's not really an Apples to Apples comparison if you just look at these quote unquote non reasoning AIS then Gemini 2.

0 flash as well as Gemini 1.